The Misunderstanding Between Tourism Resilience and Survival: Stakeholder Perceptions and Policy Effectiveness in Ecuador During the COVID-19 Pandemic Crisis

Abstract

Tourism has proven to be highly vulnerable to external disruptions, particularly in communities with low levels of tourism development. In this context, this study examines residents’ attitudes towards tourism during the COVID-19 pandemic and assesses the impact of public and private initiatives in the Cajas Massif Biosphere Area (CMBA), located in southern Ecuador. Employing a mixed-methods approach, 825 surveys were conducted alongside 25 interviews with key sector stakeholders. The objective was to determine whether these attitudes reflect genuine resilience or merely a survival strategy in response to the crisis. The findings indicate that, despite some collective efforts and mitigation plans, the primary focus remained on short-term income preservation, while government policies prioritised tourism promotion over addressing structural needs, ultimately proving inadequate for tourism recovery. This scenario placed the burden of adaptation on residents, with expressions of solidarity that, however, diminished as the crisis subsided. This study concludes that reactive measures may be mistaken for genuine resilience, highlighting the need for comprehensive policies and more equitable stakeholder participation to strengthen social cohesion and ensure the viability of tourism in the face of future crises.

1. Introduction

In general, one can summarize roughly the Ecuadorian situation during COVID-19 as a major crisis as follows. The COVID-19 pandemic exposed the vulnerability of tourism to external disruptions [1], generating profound uncertainty that led various industry stakeholders to respond in heterogeneous ways [2]. The responses of tourism stakeholders ranged from individual actions to the formation of coalitions based on shared interests. On the one hand, policymakers reinforced conventional decisions focused on tourism promotion [3]. On the other hand, private enterprises sought to keep their investments afloat [3,4]. As for community residents, their attitudes were primarily linked to the economic benefits derived from tourism [5,6,7]. However, the resilience narrative widely promoted during the pandemic framed resistance and adaptation as virtues in themselves, shifting responsibility for overcoming the crisis to individuals and communities [8]. Power structures, public policies, and economic models that perpetuate inequalities and vulnerabilities remained intact. Consequently, the notion was promoted that if communities “adapted” and “resisted”, they would overcome the challenges of COVID-19. This perspective ignored the deep asymmetries of power and systemic issues that had rendered these communities vulnerable in the first place.

The aim of this study is to determine whether residents’ attitudes towards tourism during the COVID-19 pandemic reflect resilience or mere survival. Hypothetically, the changes triggered by the pandemic may have been either short-term survival responses or part of a longer-term adaptive process, known as resilience.

The previous literature has paid increasing attention to the consequences of the pandemic for tourism on a global scale [9,10,11]; however, a specific gap remains in the in-depth exploration of how this crisis has affected residents’ attitudes in emerging destinations with low levels of tourism development [12,13,14,15,16,17,18,19]. Many studies are limited to isolated measurements, without linking their findings to a holistic analysis that would allow for a deeper understanding of the complex interplay between local community perceptions, the adoption of survival strategies, and the long-term implications for tourism resilience. Furthermore, there remains a need for methodologies that combine quantitative and qualitative approaches, as most studies tend to rely on a single method (usually surveys), without delving into the underlying dynamics that lie beneath the figures and percentages [10,20].

This research addresses that gap through a mixed-methods strategy. On the one hand, large-scale surveys of residents enable the representative measurement of general attitudes and perceptions towards tourism in a context of crisis. On the other, semi-structured interviews with key stakeholders (businesses, public bodies, and local leaders) provide qualitative insight into mechanisms of coordination, conflict, and adaptation that are not evident in purely quantitative studies. In addition, this study explores the distinction between resilience and survival from the perspective of residents in Global South communities, as well as from that of the private sector and policymakers. Tourism initiatives in these contexts often struggle to contribute effectively to livelihoods due to underlying structural development issues. This dual approach makes it possible to better understand how highly vulnerable communities respond to the abrupt disruption of tourism flows, and how such responses may shape—or hinder—the transformation of the destination in the future.

Accordingly, the central contribution of this study lies in (1) providing empirical evidence of local responses to large-scale crises in contexts with limited tourism infrastructure; (2) simultaneously examining both the micro perspective (residents’ representations, expectations, and fears) and the views of stakeholders with decision-making or planning influence; and (3) linking these perspectives to a broader debate on the need to rethink tourism growth and sustainability in times of heightened uncertainty. In doing so, this study advances a more comprehensive understanding of the social and economic consequences of the crisis in emerging destinations and proposes a reflective framework that encourages the incorporation of community resilience as a cross-cutting axis in the design of tourism policies and strategies.

This study focuses on eight different communities—rural, urban, and peri-urban—located within the Cajas Massif Biosphere Area (CMBA) in southern Ecuador. Three of these communities belong to the Coastal region and five to the Andean highlands. The selection criteria included the presence of existing tourism enterprises within the biosphere area and the willingness of key stakeholders to collaborate with the research team. Additionally, communities with varying levels of tourism development were selected, including emerging destinations, tourist corridors, and complementary attractions.

2. Literature Review

2.1. Attitudes Towards Tourism

Studies on attitudes towards tourism examine how residents and tourists perceive and react to tourism development and its associated impacts. Despite the abundance of research in this field, many studies have been criticised for their limited replicability beyond their specific contexts, making it challenging to generalise findings to other tourist destinations [9,10,11,21,22]. Attitudes can vary significantly depending on the type of destination [23], economic, sociocultural, and environmental factors [5,24,25,26,27], the level of participation [28], the destination’s development stage and future vision [29], perceptions of equity [30], prior tourism experiences, and phenomena affecting tourism [15], among others. Understanding residents’ attitudes towards tourism is critical, as these attitudes can positively or negatively influence local development through tourism [31,32]. Traditionally, the tourism sector has comprised public entities involved in tourism management and promotion, alongside private businesses operating within the tourism industry [33]. However, residents’ perspectives are vital for balanced tourism development, benefiting both the local community and visitors [12].

The study of residents’ attitudes towards tourism has been approached through various theories, which have generated diverse constructs and perspectives on how residents and local communities evaluate and respond to tourism development. In this context, Social Exchange Theory (SET) has been particularly influential, positing that residents’ attitudes are shaped by evaluating the perceived costs and benefits of tourism, with positive perceptions being more likely if benefits outweigh costs [34,35]. Meanwhile, Social Capital Theory (SCT) emphasises the role of social relationships, trust, and community norms in forming attitudes that promote tourism activity, fostering cooperation and local development [36,37]. The Theory of Planned Behaviour (TPB) provides a framework for predicting residents’ behaviour towards tourism through attitudes, subjective norms, and perceived behavioural control [38]. This theory argues that these factors influence residents’ support for tourism development in a given territory. Similarly, the Theory of Social Distance (TSD) has been used to understand residents’ attitudes towards tourism and their impact on social distance from tourists, shaping how residents interact with and accept visitors in their community [39]. Finally, Self-Perception Theory (SPT) posits that residents’ own travel experiences can influence their attitudes towards tourism [40]. Individuals form attitudes in line with their own behaviour and draw on attitudes that may have fuelled such behaviour.

Although there is no consensus on the variables defining host communities’ supportive or resistant attitudes towards tourism, they can vary depending on tourist behaviour, tourist density, and residents’ perceived tourism development levels [41]. Gursoy et al. [42] argue that attitudes towards tourism can stem from residents’ relationships with governments, community leaders, local businesses, academia, and other stakeholders concerned with residents’ opposition to tourism. According to Lindberg et al. [25], residents’ attitudes towards tourism are often linked to perceptions of economic benefits and socio-cultural factors within the community, where economic expectations tend to better predict attitudes than potential social drawbacks. In contrast, Presenza et al. [35] assert that attitudes towards tourism are shaped by perceptions of both positive impacts (e.g., increased employment and improved infrastructure) and negative impacts (e.g., increased congestion and adverse environmental effects) on local communities. These attitudes may also be associated with perceptions of quality of life and sustainability, integrated into long-term planning and community perception [43,44].

The literature on attitudes towards tourism has adopted diverse perspectives and approaches. Regarding the COVID-19 pandemic, Kamata [15] concluded that the pandemic influenced attitudes towards tourism, highlighting residents’ dilemmas between supporting the local economy and personal health concerns due to tourist interactions. Tse and Tung [45] used implicit association tests to explore how residents’ unconscious stereotypes affect their emotions and behaviours towards tourists, finding that positive implicit stereotypes are associated with positive emotions and behaviours. Torres et al. [46] developed a model to predict which consumers are more likely to engage in travel-related activities despite the challenges posed by global pandemics. Shareef et al. [47] identified reasons for changes in human psychology towards tourism during the COVID-19 pandemic to develop an attitude–behaviour model. Li et al. [16] demonstrated that residents perceive policy measures as more effective when their positive outcomes are highlighted. Additionally, residents show a greater willingness to support the mitigation of social costs through indirect means—such as receiving state aid or subsidies—rather than through direct deductions from their incomes. Guo et al. [48] found that factors such as past travel experiences, planned behaviours, perceived barriers, and the resilience of the destination significantly influence travel intentions in the post-COVID-19 period. Blackie et al. [49] argued that during COVID-19, residents were willing to accept certain inconveniences to economically benefit from the tourism industry. Erul et al. [12] developed a value–attitude–behaviour model regarding residents’ support for tourism amidst the pandemic, reflecting that residents’ valuation of tourists plays a vital role in supporting tourism.

The attitude of actors such as entrepreneurs, local communities, and public institutions towards tourism can be understood as a multidimensional construct, encompassing cognitive, affective, and behavioural components, all of which are embedded in the dynamics of each group [50,51,52]. The cognitive dimension refers to the evaluations each actor makes regarding the economic benefits, sociocultural impacts, and environmental consequences of tourism [53]. The affective dimension captures the emotions and sentiments—such as pride, satisfaction, or concern—arising from tourism development and its influence on community life and the surrounding environment [54,55]. The behavioural dimension is reflected in the actions taken by these actors, such as promoting tourism initiatives, participating in management processes, or implementing regulatory measures [56,57]. These components interact with broader contextual factors, including shared values, prior experiences, and social norms, shaping more or less favourable or resistant attitudes. Ultimately, these attitudes inform decision-making and guide practical actions concerning tourism development and governance.

2.2. Challenges of the COVID-19 Pandemic for Tourism Activity

The pandemic has posed significant challenges to tourism activities and the resilience of tourist destinations [7,58]. It remains uncertain whether the pandemic has marked a fundamental shift in tourism as a whole, and even more so for communities reliant on tourism for their livelihoods [59]. However, the pandemic has at least revitalised debates on the necessity of rethinking tourism and the balance between growth and sustainable development. Indeed, tourism is highly susceptible to crises, and tourism dependency can exacerbate vulnerabilities [60]. The instability of tourist flows during the pandemic has meant that destination resources and characteristics have become secondary to the cohesion of host communities [61].

The COVID-19 pandemic generated negative perceptions among residents, affecting their support for tourism [18]. However, this persistent threat has also led to increased responsibility and engagement among stakeholders in the tourism sector [62]. For instance, Erul et al. [12] found that residents’ hospitable attitudes were crucial for tourism support. Similarly, Qiu et al. [63] demonstrated residents’ willingness to finance risk mitigation measures during the pandemic. In this context, decision-makers have a significant impact on residents’ attitudes towards tourism. Chou Wong and Wai Lai [64] indicate that residents’ trust in their governments was influenced by the effectiveness of policy responses and the transparency of information dissemination.

For Kamata [15], COVID-19 presented a dilemma between sustaining the local economy and apprehension about welcoming tourists. However, once informed, residents, businesses, and destinations managed their interactions with visitors effectively [65]. This underscores the importance of various factors, such as engagement or aversion [28], well-being or tensions [66], the equitable distribution of benefits [30], economic perspectives [24], respect for nature and culture [26,27], the fostering of social relationships [35], the future of tourism [29], COVID-19 and tourism [15], and governance and risk management in tourism planning [17].

Moreover, several recent studies have documented how certain tourism sector stakeholders implemented adaptive strategies during the pandemic, with particular emphasis on the use of emerging technologies such as Virtual Reality (VR). In this regard, Sousa et al. [67,68] analysed how VR was employed both to maintain engagement with visitors through remote immersive experiences and to enhance on-site visits once tourism activity resumed. These initiatives helped sustain tourist interest, improve the user experience, and position companies as innovative actors, even amidst high levels of uncertainty. The studies identified multiple factors influencing the adoption of VR, including organisational dimensions (willingness to invest and the impact of COVID-19), technological aspects (perceived usefulness and expected return), and environmental factors (competitive pressure and institutional support), all framed within the Technology–Organisation–Environment (TOE) model [68]. However, significant barriers were also reported, such as lack of technical knowledge, high implementation costs, and limited infrastructure—particularly among small and medium-sized enterprises [67].

The recovery of tourism systems presents a significant challenge, partly because organisations tend to prioritise their own objectives [69]. However, the crisis has highlighted the growing focus on resilience as a key capacity for adapting to new scenarios [65,70]. Therefore, rather than focusing solely on recovery, it is essential to examine the underlying causes of vulnerability [2]. Returning to “our previous state” necessitates a critical reflection on who constitutes this “we” and whether that previous state is truly desirable [71].

2.3. Resilience or Survival Instincts?

The literature indicates that the concept of “tourism resilience” lacks a unified definition [17,69,72]. An ideal understanding should encompass multiple dimensions, including community, destination, and organisational resilience [17]. This semantic plurality can create confusion when linking resilience to tourism, particularly when the core of the concept revolves around adaptation and recovery in the face of adversity [73].

Within the tourism system, resilience involves interactions and processes at local, regional, and global scales, resulting in varying responses to adversity [17]. Hall [72] argues that tourist destinations are subcomponents of a broader tourism system that is interconnected with socio-economic changes. Resilience refers to the ability of communities and individual residents to adapt and thrive amid uncertainty, which is crucial during crises [17,74,75]. The capacity for direct change is framed by dynamic structures and relationships [69]. From a business perspective, Prayag [17] contends that resilience is shaped by political and economic dynamics. At the organisational level, factors such as staff adaptability and supply chain flexibility come into play; however, they are also influenced by individual entrepreneurial traits and the broader business context.

Resilience in the tourism sector is particularly challenging in the context of economic and social preparedness for crises [76,77]. In such scenarios, the measures adopted often tend to be unsustainable and driven by desperation. Less developed regions are generally less resilient due to factors such as economic development levels, financial resource availability, and administrative capacity [78]. Furthermore, poor coordination among tourism stakeholders can result in slow and fragile responses to crises [58,79].

Additionally, resilience is influenced by the intrinsic characteristics of the tourism system and requires effective leadership to manage change [17]. Thus, proactively addressing vulnerability factors in tourism is crucial [72]. Evans and Reid [8] define vulnerability as a system’s inherent quality that predisposes it to adverse effects in the face of risk. Clark et al. [80] distinguish between exposure to risk in terms of enduring change without alteration (survival) and the ability to manage and adapt to changes (resilience). Indeed, the concept of resilience in tourism is far from straightforward, particularly when its application remains confined to metaphorical descriptions based on normative and positive assumptions [81].

Although interrelated, survival and resilience are distinct concepts with key nuances. Survival is defined as the ability to persist despite adverse conditions and is often an automatic response to immediate threats, driven by fundamental biological and psychological mechanisms [8,82,83]. By contrast, resilience entails a more prolonged and complex process of adaptation and recovery in response to various adversities, encompassing psychological, economic, social, and cultural dimensions [17,84]. Resilience often operates within a collective or community framework [85]. While the survival instinct is reactive and short-term focused, resilience represents a continuous adaptive cycle that may include phases of recovery and growth [81].

3. Description of the Study Area and Methodology

3.1. Study Area

This study examined eight significant cases within an Ecuadorian–Belgian academic project in the Cajas Massif Biosphere Area, in southern Ecuador (Figure 1) [86]. In the Andean region, this study incorporates peri-urban cases such as Baños; rural cases like Sayausí and Migüir; and urban–rural interaction cases, including the Austro Agroecological Producers and the Hat Museum in Cuenca. In the coastal zone near Guayaquil, rural cases from the Naranjal Cluster, Tsuer Entsa, and 6 de Julio were examined. The selection of these case studies was carried out through a participatory process lasting approximately two months. This involved approaching communities, presenting the project objectives, and assigning shared responsibilities. The inclusion criteria were defined based on both operational feasibility and the willingness of local actors to collaborate. Specifically, the selected territories met the following requirements: (1) being located within the area of influence of the Cajas Massif Biosphere Area (CMBA); (2) having initiated tourism-related activities; (3) expressing interest in being part of the project; (4) allowing students to conduct fieldwork; and (5) formally agreeing to sign a cooperation agreement. It is worth noting that the final three criteria were the most difficult to meet, as they required a level of commitment, institutional openness, and coordination among actors not always present in the initially considered locations [87]. This selection aimed to ensure suitable conditions for participatory engagement and the rigorous systematisation of information in real-world contexts.

The parish of Baños, located approximately 8 km from the city of Cuenca, is a nationally recognised destination in the field of health and wellness tourism. Its thermal springs, spa services, and natural surroundings attract visitors seeking relaxation experiences and therapeutic treatments. Its proximity to Cuenca—an internationally significant cultural destination—reinforces its position as a strategic site for both local and national tourism.

The parish of Sayausí, situated to the northwest of Cuenca, is notable for its proximity to Cajas National Park, making it an important hub for ecotourism and nature-based tourism. Activities such as hiking, wildlife and flora observation, and agrotourism are integral to its tourism offer. Its privileged landscape has encouraged the development of local enterprises with a sustainable focus.

Miguir, strategically located along the road connecting the provinces of Azuay and Guayas, plays a key role in the transit of travellers and the trade of agricultural products. A significant portion of its territory lies within Cajas National Park, allowing it to combine gastronomic tourism—particularly centred on trout production and consumption—with nature-based tourism.

The Agroecological Network of the Austro (Red Agroecológica del Austro) brings together farming families and agricultural ventures committed to organic and sustainable production in the southern region of Ecuador. This network facilitates collective marketing and the distribution of essential goods—particularly significant during times of crisis, such as the pandemic—by reducing intermediaries and strengthening the local economy.

The Hat Museum (Museo del Sombrero), located in Cuenca’s historic centre, showcases the traditional craft of weaving “the paja toquilla hat”, recognised by UNESCO as part of the Intangible Cultural Heritage of Humanity. Beyond functioning as an exhibition space, the museum promotes the preservation and transmission of artisanal techniques that are integral to Ecuadorian cultural identity.

Naranjal Cluster, located in Ecuador’s coastal region, approximately 70 km from the city of Guayaquil, stands out as an agricultural and commercial corridor between the Coast and the Highlands—a position that has historically facilitated the movement of goods and people. The canton hosts a protected area with high biodiversity in terms of flora and fauna. This refuge has become an ecotourism hotspot offering birdwatching, canoe tours through the mangroves, and other nature-based activities. Additionally, there are community-based tourism experiences focused on artisanal fishing and traditional gastronomy, which make use of the area’s rich aquatic resources—such as crab and shrimp—and its proximity to rivers and estuaries.

The community of Tsuer Entsa is inhabited by families of the Shuar nationality, who preserve their ancestral customs, language, and practises. The Shuar identity is expressed through their close relationship with nature and the sustainable use of water and geothermal resources. The community’s main attraction is its thermal springs, which appeal both to local tourists and to those seeking health and wellness experiences in natural environments.

The community of 6 de Julio is located in an area of wetlands and mangroves, and its local economy combines traditional subsistence activities—particularly the harvesting of red crab—with an emerging model of community-based tourism focused on environmental and cultural interpretation. The tourism offer includes immersive experiences in which visitors can learn about the crab-harvesting process, alongside dolphin, bird, plant, and wildlife observation within the mangrove ecosystem. These initiatives have become a sustainable and educational alternative that connects tourists with local knowledge and ecosystem conservation.

The Tourist Corridors (TC), Sayausí and Migüir, are located along the Cuenca-Guayaquil highway and feature tourism infrastructure that complements local productive activities. The Complementary Attractions (CA) include the Hat Museum, recognised by UNESCO, and the Austro Agroecological Producers, both of which connect urban and rural areas through their product offerings. Although still emerging, the Tourist Destinations (TD) exhibit high levels of interaction among stakeholders and an increasing reliance on tourism. These destinations are characterised by their cultural and natural elements, such as indigenous communities and diverse landscapes.

For analytical purposes, this study classified the localities into three operational categories: “Tourist Destinations”, “Tourism Corridors”, and “Complementary Attractions”. This categorisation does not correspond to a formally established typology in the academic literature but was developed based on empirical observation of each locality’s level of tourism development, socio-economic dynamics, and functional role within the regional tourism system. “Tourist Destinations” are characterised by a consolidated offer (infrastructure, services, and promotion) and a high influx of visitors, resulting in more intense interaction with tourism and the generation of both opportunities and challenges for residents and managers. “Tourism Corridors” refer to areas with frequent transit, used as stopover or passage points, where the tourism offer caters to functional needs related to mobility and connectivity. Lastly, “Complementary Attractions” designate specific sites which, although not principal destinations, enhance the visitor experience and may represent a source of cultural and economic value for local communities. This distinction categorisation allowed for a more accurate identification of the particularities of each context and enabled the analysis of how perceptions of tourism vary according to the type of interaction established between local actors and visitors.

3.2. Methodology

The objectives of this study were to provide empirical evidence of local responses to large-scale crises in contexts with limited tourism infrastructure; to simultaneously examine the representations, expectations, and concerns of residents alongside the perspectives of actors involved in decision-making and planning; and to link these perspectives to the broader debate on the need to rethink tourism growth and sustainability in times of high uncertainty.

A sequential mixed-methods observational design was employed [88]. The quantitative phase was conducted through anonymous and voluntary surveys administered between July and August 2021 using mobile devices and the Kobo Toolbox application. The field team consisted of three supervisors and fourteen previously trained students, who ensured the proper administration of the questionnaire.

The sample consisted of residents from eight localities selected as case studies (see Figure 1). In each locality, 100 participants were randomly selected, including both individuals involved in tourism activities and those with no direct connection to the sector. This strategy followed a non-probabilistic convenience sampling approach, using fixed quotas per geographical unit in order to ensure a homogeneous treatment of attitudes towards tourism, despite population differences between the communities [89]. Due to the difficulty of obtaining exhaustive and updated lists of residents with contact information, previous fieldwork in each community was used to contact individuals linked to tourism or close to its activities. Following a rigorous data cleaning process, the final sample included 825 individuals, as detailed in

Table 1. Stratified sample (n = 825).

Study Case	Population (Except Children)	Sample Size	Sample to Population Ratio
Baños (D)	9266	101	12%
Sayausí (TC)	4475	105	31%
Migüir (TC)	160	100	98%
Agro-ecological producers of the Austro (CA)	4071	100	28%
Tsuer Entsa (D)	150	100	91%
Naranjal Cluster (D)	560	110	60%
6 de Julio (D)	949	109	73%
Hat Museum (CA)	4863	100	91%
Total	24,494	825

, and a non-proportional stratification was applied to appropriately reflect the heterogeneity of the cases included in this study.

Accordingly, residents were surveyed given their size and diversity, which allowed for the collection of representative and quantifiable data on general attitudes towards tourism. As those who most directly and regularly experience tourism’s sociocultural and economic impacts, their responses to structured questionnaires enabled a systematic analysis of acceptance levels, perceived benefits, and potential tensions related to tourism development.

3.3. Data Collection Instrument

The questionnaire was structured into three sections. The first included control and supervision variables (date, time, names of the interviewer and supervisor, case study, and informed consent statement). The second section comprised 36 statements distributed across nine macro-variables: involvement or aversion [28,42], well-being or tensions [66,90], equitable benefit distribution [30], economic issues derived from tourism [24], respect for nature [27,91], respect for culture [26], social relationship-building [35], the future of tourism [29], and COVID-19 and tourism [15]. A seven-point Likert scale was used (from 1, “strongly disagree”, to 7, “strongly agree”). The third section collected socio-demographic variables (see

Table 2. Description of the sample: demographic data.

Demographic Attribute	Category	Percent of Total
Gender	Male	43
	Female	56.9
	Not specified	0.1
Workplace	Inside the community	80
	Outside the community	20
Income dependence	Public sector	24.1
	Private sector	5
	Own business/entrepreneurship	58.2
	No income	12.1
	No reply	0.6
Link to tourism	Direct	31.3
	Indirect or not link to tourism	68.7
Qualification	Primary school	41.3
	Secondary school	42.7
	University	10.1
	Postgraduate	0.8
	No studies	50.1
Age	Min	18
	Average	38.7
	Max	83

). The “Don’t know” option was excluded after assessing the internal reliability of the scale using Cronbach’s alpha [92], which yielded values of 0.775 (95% CI: 0.743–0.805) in a 50% random subsample and 0.759 (95% CI: 0.735–0.782) in the total dataset (N = 825), indicating acceptable internal consistency and supporting the decision to minimise non-committal responses.

3.4. Data Analysis

Data analysis was conducted using R Studio version 4.1.2. Initially, descriptive analyses were conducted for both socio-demographic variables and responses to the 36 statements. Given that mid-range scores (3, 4, and 5) could indicate indecision, the analysis focused on extreme responses, specifically high levels of agreement (scores 6 and 7) and disagreement (scores 1 and 2). The “I don’t know” option was excluded from the questionnaire to minimise non-committal responses.

The analysis began with an exploratory data analysis (EDA) by calculating the median of each statement. This approach aimed to reduce the subjectivity of Likert scales and identify trends towards lower (Disagree) and higher (Agree) values. Subsequently, multiple stages of statistical analysis and predictive modelling were implemented. The methodological sequence included: Principal Component Analysis (PCA), cluster segmentation using K-means (K-means is a statistical method used to partition data into distinct clusters based on similarity, thereby enabling the identification of patterns or underlying structures within large datasets). And decision tree (A decision tree is a statistical modelling technique that uses a tree-like structure to represent decisions and their possible consequences, allowing for the classification or prediction of outcomes based on input variables) construction. For the PCA, responses were standardised (z-score) before applying PCA to the standardised dataset to reduce dimensionality and extract key components that explain the greatest variance in residents’ attitudes. The Varimax rotation method was used to ensure maximum variance of the data. The Kaiser criterion was applied, selecting components with eigenvalues greater than 1, which led to the identification of seven principal components. Based on the PCA results, the elbow method was employed to determine the optimal number and size of clusters, resulting in four clusters (190, 117, 262, 256). Using the principal components extracted from the PCA as input variables, the K-means algorithm was applied to segment respondents’ answers into the four identified clusters.

A decision tree was then constructed to predict the clusters assigned by the K-means algorithm within a dataset containing relevant segmentation variables. To ensure reproducibility, a random partitioning of the dataset into training and testing sets was performed. Specifically, 70% of the data was randomly selected for the training set, while the remaining 30% was used as the test set. The decision tree was subsequently used to predict cluster assignments in the test set. At each decision tree node, the mean cluster value was computed for all observations satisfying the conditions up to that node. This mean cluster value was calculated across all four identified clusters (1, 2, 3, and 4), allowing for intermediate nodes to represent averages between clusters. To assess the model’s performance, key metrics such as accuracy (Accuracy refers to the proportion of correct predictions made by the decision tree), confidence intervals (Confidence intervals indicate the range within which the true accuracy is likely to fall), p-values (p-values assess the statistical significance of the model’s results.), and the Kappa index (Kappa index measures the agreement between predicted and actual cluster assignments beyond chance) were evaluated. Additionally, stability and robustness were examined through simple cross-validation (Simple cross-validation is a method where the dataset is randomly split into a training set and a test set to evaluate the model’s ability to generalise to unseen data), by splitting the data into a training set (70%) and a test set (30%). The model was trained on the 70% training data, and its performance was then validated using the remaining 30% of the dataset, which had not been used in training. This approach ensured that the model was not overfitted to the training data and could generalise well to unseen data. It was crucial for verifying whether the model remained consistent and robust across different data partitions. In practice, this validation method helped confirm that the results were not dependent on a single data partition but remained stable across different samples of the dataset [89]. By changing the training and test data, variations in the decision tree structure were observed. If the resulting trees were similar, the model was deemed robust. However, if significant differences emerged, this suggested that the model was sensitive to data selection and might require further adjustments.

3.5. Qualitative Data Collection and Analysis

The initial implementation of surveys ensured a robust quantitative foundation, which guided the selection of topics for interviews, optimising the qualitative approach and preventing a dispersed or redundant exploration. The interviews were designed to delve deeper into critical areas previously identified through quantitative data analysis (see Appendix B) [88]. An exploratory approach was adopted, employing semi-structured interviews. The participants included tourism authorities (5), business owners (4), and local leaders (16), selected based on convenience sampling (

Table 3. Characteristics of the semi-structured interviews.

Sector	Code	Interviewed	Gender	Age	Format	Length
Public sector	PS01	Local government	Female	39	Virtual	0:41:24
	PS02	National government	Female	58	Virtual	0:44:45
	PS03	Local government	Male	51	Virtual	0:47:01
	PS04	Local government	Male	35	Virtual	0:31:24
	PS05	Local government	Female	46	Virtual	1:07:26
Private sector	BS01	Businessman	Male	54	Face to face	0:33:20
	BS02	Businessman	Male	48	Face to face	0:31:06
	BS03	Businessman	Female	26	Virtual	0:36:09
	BS04	Businessman	Male	32	Virtual	0:41:16
Local communities	LC01	Local	Female	31	Face to face	0:26:16
	LC02	Local	Female	24	Face to face	0:58:51
	LC03	Local	Female	50	Face to face	1:14:49
	LC04	Local	Male	38	Face to face	0:28:15
	LC05	Local	Male	41	Face to face	0:27:46
	LC06	Local	Female	34	Face to face	0:21:44
	LC07	Local	Male	34	Face to face	0:20:12
	LC08	Local	Male	71	Face to face	0:21:35
	LC09	Local	Male	44	Face to face	0:26:55
	LC10	Local	Male	45	Face to face	0:20:59
	LC11	Local	Female	44	Face to face	0:25:29
	LC12	Local	Female	31	Face to face	0:21:40
	LC13	Local	Male	35	Face to face	0:45:41
	LC14	Local	Female	30	Face to face	0:38:47
	LC15	Local	Male	72	Face to face	0:25:32
	LC16	Local	Male	61	Face to face	0:19:54

). Unlike the broad and diverse resident population, these actors are more focused groups with specialised knowledge or decision-making roles, making interviews a more suitable method for exploring their perspectives, motivations, and challenges in tourism management. All of this aimed to gain a deeper understanding of governance processes, planning, and cross-sector coordination, thereby complementing the findings from the quantitative survey. In total, 25 interviews were conducted between November 2021 and March 2022, of which 18 were face-to-face and 7 were carried out via Zoom meetings, with prior informed consent. The duration of the interviews ranged from 20 min to 1 h and 15 min.

Qualitative data were analysed through a systematic coding process that enabled an in-depth interpretation of the interview content within its social and territorial context. All interviews were recorded and transcribed and subsequently translated from Spanish into English in a highly literal manner to preserve the integrity of the captured dialogues. Three researchers conducted a detailed reading of the transcripts, applying an inductive method to identify emerging themes. This process led to the creation of a set of codes, which were systematised using the qualitative analysis software Atlas.ti (version 23). In addition, discourse analysis was employed to examine the representations, metaphors, and interpretative frameworks present in the responses, thereby facilitating a deeper understanding of underlying meanings [93]. This qualitative approach enriched the interpretation of perceived motivations and challenges, complementing and contextualising the results obtained in the quantitative phase [88].

4. Results

4.1. The Statements Assessed: Some Figures

Figure 2 displays 20 statements with strong opinions categorised according to the median of their scores on the Likert scale. The chart is divided into potentially unaffected factors (depicted on the right with dashed lines) and those clearly influenced (on the left) by the pandemic. For instance, the perception that “tourism contributes to species conservation” (Q32) could be stable regardless of the pandemic, in contrast to the statement “tourism will be the predominant activity in the future” (Q17).

The right-hand side of the chart shows a positive attitude towards tourism, except for question 31, which reflects a sense of realism. The left-hand side of the chart contains statements sensitive to the “pandemic”, as tourism, although underdeveloped, came to a halt and impacted many people. In this case, 45% of the respondents claimed to engage in local tourism (Q1); thus, opinions about tourism were generally positive. However, a sense of realism emerges, particularly encapsulated by the statement “The COVID-19 pandemic has shown that we cannot depend on tourism as a source of income in my community” (Q9). This contradicts the equally significant statement “In the future, tourism will be the main activity in my community” (Q17). This reveals that, despite the weaknesses of tourism, residents still hope to improve their lives.

4.2. Principal Component Analysis (PCA)

As outlined in the methodology chapter, we aimed to explore broader dimensions and associations, going beyond individual variables (statements) through the application of Principal Component Analysis (PCA). When considering all cases and respondents, the PCA identifies five components that account for 41.6% of the total variance in the data (

Table 4. Attitudes towards tourism (PCA ordinate loadings ≥ 0.5 and ≤−0.5).

Variable	Dimension 1	Dimension 2	Dimension 3	Dimension 4	Dimension 5
Improve trust (Q7)	0.68
Help women (Q8)	0.68
Help local prod. activities (Q6)	0.67
Conflicts from tourism (Q2)	−0.54
Organised community (Q29)	0.54
Employment/business tourism (Q4)	0.51
Work in tourism (Q28)		0.75
Active participation in tourism (Q1)		0.73
Income from tourism (Q13)		0.63
Quality of life (Q20)		0.62
Insecurity (Q34)			0.76
Trash (Q23)			0.71
Water pollution (Q14)			0.69
Prostitution/Alcohol/Drugs (Q25)			0.63
Vehicular disorder (Q5)			0.61
Cost of living increase (Q31)				0.63
Tourism as a main activity (Q17)				0.55
Cannot depend on tourism (Q9)					0.59
Eigenvalues	3.88	3.22	3.09	1.82	1.69
VAR	20.6%	8.1%	5.2%	3.9%	3.8%

). While this result is not entirely satisfactory as a means of summarising dimensions, it suggests the presence of multiple independent variables, which aligns with the fact that the correlation analysis did not reveal many high correlations among the statements (Annex 1). Below, we present the underlying variables identified through the PCA:

4.2.1. Dimension 1: Economic Mutualism

This dimension brings together variables related to trust (Q7), women’s empowerment (Q8), support for local productive activities (Q6), community organisation (Q29), the perception of tourism as a source of employment (Q4), and, inversely, tourism-related conflicts (Q2). These variables suggest a form of community cohesion centred on the expected economic benefits of tourism. This pattern aligns with the findings of Strzelecka et al. [94], who argue that tourism can strengthen the social fabric when benefits are distributed equitably. However, experiences during the pandemic revealed that such cohesion can be fragile and more rooted in economic expectations than in consolidated organisational processes. In fact, Lee and Jan [31] warn that initial economic interest may act as a mobilisation driver but can also generate tensions if the benefits are neither sustainable nor inclusive.

4.2.2. Dimension 2: Socioeconomic Participation in Tourism

The associated variables (Q1, Q13, Q20, Q28) reflect high levels of local population involvement in tourism activities, primarily driven by income generation and improvements in quality of life. This type of engagement reinforces economic dependence on tourism, which may translate into a generally positive perception, particularly during periods of sector growth. This phenomenon has been widely documented in rural economies that rely on tourism as their main income source [95], which increases vulnerability to external crises [1], as evidenced during the COVID-19 pandemic. Although such participation can enhance living conditions in the short term, it may also displace other key dimensions such as sustainability, economic diversification, and equitable distribution of benefits [96].

4.2.3. Dimension 3: Deferral of Tourism’s Social and Environmental Costs

This dimension groups variables related to social and environmental issues: traffic congestion (Q5), water pollution (Q14), littering (Q23), the presence of prostitution, drugs and alcohol (Q25), and insecurity (Q34). Their clustering indicates that, while these impacts are acknowledged, they tend to be downplayed or postponed, especially during the early stages of tourism development. This type of “functional blindness” to tourism impacts has been described by Almeida García et al. [5], where enthusiasm for immediate economic benefits can overshadow long-term planning. Such omission may lead to a gradual deterioration of quality of life and the sustainability of the destination, ultimately affecting its competitiveness and reputation.

4.2.4. Dimension 4: Awareness of Tourism Development

The grouped variables (Q17, Q31) reflect a dual perception of tourism: on the one hand, as the main economic activity; on the other, as a factor that increases the cost of living. This duality suggests a growing awareness of the structural effects of tourism on the local economy and daily well-being. In this regard, Andereck [97] notes that communities may adopt an active role in reflecting on the benefits and costs of tourism, particularly when facing rapid tourism growth. This dimension could indicate a transition from an optimistic view to a more critical stance on the transformations brought about by tourism development.

4.2.5. Dimension 5: COVID-19 and Tourism—Lessons Learned, Learning Forgotten

The variable (Q9) reflects the perception that communities cannot rely solely on tourism. This idea emerged strongly during the pandemic, when communities demonstrated collaboration and adaptability. However, after the reopening, there has been a tendency to return to tourism-centred growth models, leaving behind some of the lessons learned. Prayag et al. [17] warn that many so-called “resilience” strategies implemented during the COVID-19 crisis were not institutionalised but rather reactive responses to uncertainty. Genuine resilience, as Hall [72] argues, requires learning, structural adaptation, and transformation—not just temporary resistance.

4.3. Cluster Analysis Based on PCA Dimensions

Four clusters have been identified, as shown in

Table 5. Dimensions of tourism perceptions across clusters.

Dimensions	Cluster 1: Economic Pragmatists	Cluster 2: Critical Realists	Cluster 3: Survivalist Idealists	Cluster 4: Moderate Sceptics
Dimension 1: Economic mutualism	0.24	−1.76	0.50	0.11
Dimension 2: Socioeconomic participation in tourism	−0.06	−0.52	−0.72	1.02
Dimension 3: Deferring the social and environmental costs of tourism	0.20	0.06	−0.29	0.12
Dimension 4: Awareness about tourism development	−0.29	0.21	−0.05	0.17
Dimension 5: COVID-19 and tourism: Lessons, yes; learning, no	−1.27	0.11	0.52	0.36

, demonstrating moderate separation and total internal variability (24.4%) due to differences among clusters. The sum of squares reveals those observations in Cluster 2 are more tightly grouped. In contrast, Cluster 3 has the highest within-cluster sum of squares, indicating that observations in this cluster are more differentiated.

The values in the table represent standardised mean scores (z-scores), indicating how strongly or weakly each cluster aligns with the identified dimensions in relation to the total sample. In this context

• A score of 0 reflects the average perception (equal to the overall sample mean).
• Scores greater than 0 indicate a more positive perception than average.
• Scores less than 0 indicate a more negative perception than average.

For example, in Cluster 1 (Economic Pragmatists), the score of 0.24 for Dimension 1 (Economic Mutualism) suggests that individuals in this group hold a slightly more favourable perception of economic cohesion through tourism than the overall average. In contrast, Cluster 2 (Critical Realists) has a score of −1.76 in the same dimension, reflecting a significantly more critical view regarding tourism’s potential to generate economic cohesion.

4.3.1. Cluster 1: Economic Pragmatists

This cluster views tourism primarily as an economic tool (Dimension 1: +0.24) and focuses on immediate financial benefits, often downplaying social and environmental risks. While they exhibit slightly positive attitudes towards economic mutualism and the deferral of tourism’s social and environmental costs (Dimension 3: +0.20), their engagement in tourism-related decision-making is limited (Dimension 2: −0.06), and their awareness of broader development implications remains low (Dimension 4: −0.29). This indicates that, although they recognise tourism’s economic potential, they do not fully perceive longer-term challenges such as rising living costs or the need for diversification. Moreover, this group scored the lowest on the COVID-19 dimension (Dimension 5: −1.27), reflecting a preference for returning to pre-pandemic models without integrating the lessons learned. Their responses suggest a survival-oriented logic driven by short-term gains, with little indication of transformative or adaptive resilience.

4.3.2. Cluster 2: Critical Realists

This cluster demonstrates a strongly critical stance towards tourism’s economic benefits (Dimension 1: −1.76), suggesting a belief that such benefits are not equitably distributed among residents—possibly due to internal conflicts or ineffective self-organisation within CMBA communities. They also express scepticism towards socioeconomic participation in tourism (Dimension 2: −0.52), reflecting limited trust in a development model based solely on tourism. Nevertheless, they exhibit slightly above-average awareness of tourism’s broader impacts, particularly regarding social and environmental issues (Dimensions 3, 4, and 5). In particular, their awareness of development-related challenges (Dimension 4: +0.21) indicates concern about long-term risks such as rising living costs and a more cautious outlook on tourism’s role in their communities. Furthermore, their perception of the pandemic is neutral to slightly positive (Dimension 5: +0.11), suggesting that while they recognise certain lessons from COVID-19, the crisis has not fundamentally changed their views on tourism. Overall, this group tends to focus more on the structural limitations of tourism development, approaching it with critical reflection rather than optimism.

4.3.3. Cluster 3: Survivalist Idealists

This cluster combines a positive view of tourism as a key economic driver (Dimension 1: +0.50) with a critical stance towards socioeconomic participation (Dimension 2: −0.72), suggesting that while they see tourism as vital for their community, they believe the current model fails to provide fair or sufficient opportunities for all. Respondents in this group are less actively involved in tourism processes, indicating a disconnect between their aspirations and their engagement. They also hold a negative view regarding the deferral of social and environmental costs (Dimension 3: −0.29), reflecting concern for long-term impacts despite recognising short-term economic benefits. Notably, this group shows strong acceptance of pandemic-related lessons (Dimension 5: +0.52), implying that they have internalised key reflections from the COVID-19 crisis and support a transformation of tourism towards greater resilience and equity. However, their limited awareness of broader tourism development issues suggests an idealistic yet disengaged position. Overall, these residents may be better prepared to face future crises or contribute to strategic, long-term responses, even if their current involvement is limited.

4.3.4. Cluster 4: Moderate Sceptics

This group maintains a balanced perspective, showing no extreme values across any of the five dimensions. They display a notably positive attitude towards socioeconomic participation in tourism (Dimension 2: +1.02), suggesting an appreciation for tourism’s contribution to quality of life and employment opportunities. However, they are more cautious regarding economic mutualism (Dimension 1: +0.11) and the deferral of social and environmental costs (Dimension 3: +0.12), indicating some awareness of the risks and limitations of an overreliance on tourism. Their perception of tourism development is moderately positive (Dimension 4: +0.17), reflecting recognition of both the benefits and emerging challenges, such as rising living costs. While their responses demonstrate a more neutral and less polarised stance, this could also indicate a degree of disengagement or limited critical reflection on tourism-related issues in their communities. Overall, the group appears moderately engaged but potentially underprepared to respond to deeper structural transformations in the tourism sector.

4.4. Decision Tree Analysis

The decision tree analysis reveals key divisions in residents’ perceptions of tourism. The root node displays a mean value of 2.7, and the model shows moderate performance overall (The model demonstrates moderate accuracy (65.6%) with a confidence interval of 0.593–0.7149. It performs significantly better than random prediction (p < 2.2 × 10⁻¹⁶), and the Kappa value (0.5335) suggests moderate agreement.). The model suggests that most residents belong to Cluster 2 (Critical Realists) or Cluster 3 (Survivalist Idealists). Notably, 78% of residents are open to tourism activities promoted within their community (Q19 ≥ 6; mean value: 3). This indicates a predominantly positive attitude towards tourism, though it aligns more closely with Cluster 3, which reflects a survival-oriented approach rather than genuine resilience. Within this branch, 65% of residents do not primarily work in tourism (Q28 < 6; mean value: 2.8), indicating a moderate attitude towards tourism, which suggests a survivalist approach rather than full integration into the sector. Among this group, 52% do not believe that tourism has significantly benefited women in terms of employment or business opportunities (Q08 < 5; mean value: 3). This suggests that residents—primarily from Cluster 3—perceive that tourism has not fostered equity or sustained economic impact, revealing a lack of resilience in terms of equitable development.

A five percent (5%) of residents have a neutral or negative perception regarding post-pandemic tourism dependency (Q09 < 4; mean value: 1.9), reflecting scepticism about tourism’s ability to sustain the local economy in the long term. This aligns with Cluster 2 (Critical Realists), who adopt a critical stance and perceive tourism primarily as a temporary survival resource rather than a sustainable economic pillar. Additionally, 4% of these residents believe that tourism should be supplemented with other productive activities (Q26 < 7; mean value: 2.2), reinforcing the need for economic diversification. This further supports the notion that these groups do not exhibit clear signs of economic resilience but rather demonstrate partial dependence on tourism as a survival strategy.

In contrast, 45% of residents believe that exclusive dependence on tourism is not feasible (Q26 ≥ 5; mean value: 3.2), reflecting a more critical awareness of the limitations of tourism as the sole source of income. However, this group continues to emphasise the need for additional productive activities, indicating a long-term resilient and adaptive outlook. Moreover, 25% of residents do not actively participate in community tourism activities (Q01 < 5; mean value: 2.9), while 20% are involved in the tourism sector. This suggests that while there is openness towards tourism, its impact has not been broad or inclusive enough to foster wider participation among community members.

Within this group, 16% of residents believe that it is possible to rely solely on tourism for their livelihood (Q26 ≥ 7; mean value: 3.6). This reflects an optimistic yet minority perspective, characteristic of Cluster 3 (Survivalist Idealists). However, 5% of residents adopt a more cautious attitude, indicating doubts about tourism’s viability as the sole source of income. Finally, a small percentage (2%) is fully open to tourism (Q19 ≥ 7; mean value: 4), reflecting a strongly positive attitude toward the sector, though this support remains limited in size.

Among the residents located in the branch where Q19 < 6, 22% are not open to tourism (mean value: 1.8), and more critical attitudes predominate. This subgroup aligns with Cluster 1 (Sceptics) or Cluster 2 (Critical Realists). These residents do not believe in the equitable distribution of tourism income (Q03 < 5; mean value: 1.4), reinforcing the perception that tourism has not been inclusive or equitable (see Figure 3). This indicates that survival strategies have been adopted to leverage limited resources, but without generating structural changes in the community. Furthermore, 11% of the residents within this same subgroup explicitly recognise that tourism income is not distributed fairly (Q03 ≥ 5; mean value: 2.1), demonstrating a lack of trust in tourism’s ability to foster long-term economic resilience.

Simple Cross-Validation Results from the Second Decision Tree

The key percentages between both decision trees are highly similar, with the greatest difference observed in the 37% of residents who believe that tourism has benefited women. This suggests a more favourable perception of tourism in terms of gender equity in this second tree. This could indicate that, while overall attitudes remain oriented towards survival rather than resilience, there are signs that certain sectors of the community (such as women) have begun to benefit more from tourism in this context.

The model’s accuracy is 64.37%, meaning it correctly classifies approximately two-thirds of observations. Although this is slightly lower than in the first tree, it still represents a moderate level of accuracy (see Figure 4). The second decision tree presents moderate accuracy in classifying residents’ attitudes towards tourism, showing a significant improvement over a baseline model (The model’s 95% confidence interval ranges from 58.05% to 70.34%, with a No Information Rate (NIR) of 0.3279. Its accuracy significantly surpasses this baseline, and the extremely low p-value (<2.2 × 10⁻¹⁶) confirms the model’s predictive power. The Kappa value (0.5209) indicates moderate agreement between predicted and actual values. No Information Rate (NIR) represents the accuracy that would be achieved by always predicting the most frequent class in the dataset, serving as a baseline to compare whether the model adds real predictive value). While not perfect, this reflects a satisfactory model performance in classifying residents’ attitudes towards tourism.

4.5. Contextual Insights from a Qualitative Perspective

Qualitative findings provided nuanced insights, occasionally challenging results from quantitative analyses (PCA, cluster analysis, and decision tree model). Although statistical models identified respondents’ attitudes as primarily pragmatic and economically driven, with limited learning from the pandemic experience, qualitative interviews revealed temporary collaborative practises rooted in reciprocity and solidarity. Such practises, however, quickly dissipated once lockdown restrictions eased and tourism resumed. Thus, the quantitative emphasis on economic motivations did not equate to genuine collective resilience or sustained social transformation.

The testimonies of key stakeholders affirmed the following:

“We came together with the hope that tourism would bring us better economic opportunities, but this unity was temporary. After the pandemic, individual economic interests prevailed.”

(LC01)

Furthermore, a local entrepreneur closely connected to the coastal communities stated the following:

“We united because there was no other option, but as soon as things returned to normal, we went back to business as usual.”

(BS03)

The PCA’s Dimension 5 (“COVID-19 and tourism: Lessons, yes; learning, no”) indicates that, although certain lessons were acknowledged, these did not lead to sustained behavioural or structural changes. However, qualitative evidence provides a more nuanced interpretation, identifying a phenomenon described as “ephemeral solidarity”. Interviews revealed that collaboration and mutual support were primarily instinctive and short-lived crisis responses rather than evolving into enduring governance mechanisms or participatory planning strategies. While quantitative analyses might suggest a simple return to “business as usual”, stakeholder perspectives clarify the underlying factors driving this outcome: the absence of strategic vision, the pressing need for short-term income recovery, and limited public-sector involvement in fostering genuine community resilience.

A local community leader stated the following:

“We came together with the hope that tourism would bring us better economic opportunities, but this unity was temporary. After the pandemic, individual economic interests prevailed.”

(LC01)

The PCA dimension labelled “Economic Mutualism” highlights a significant dependence on tourism as an income source. However, qualitative interviews reveal that this dependence is neither homogeneous nor consistent across communities. Although quantitative analyses categorise groups as “Survivalist Idealists” or “Moderate Sceptics”, qualitative evidence indicates that some communities effectively adopted alternative livelihoods—such as agriculture, fishing, and bartering—temporarily reducing their reliance on tourism. Conversely, other territories exhibited greater tourism dependency, becoming economically paralysed without visitors and necessitating governmental or external assistance.

While communities initially organised around tourism as a unifying economic force, this motivation can also generate internal conflicts related to benefit distribution, decision-making processes, and unmet expectations. Qualitative insights thus challenge the statistical assumption that tourism-dependent communities experienced crises uniformly. Instead, qualitative evidence emphasises that dependency levels are influenced by support networks, organisational capacities, and stakeholder commitment.

Residents stated the following:

“Tourism provides us with jobs and helps us move forward, but we depend on visitors coming. If that fails, we have no other option.”

(LC02)

“During the pandemic, there was no tourism, and that was our main activity […] so we focused on the small fields we had around here. We had crops planted on our plots up the verde (hill—greens), yuquitas (cassava) here and there. We also hunted wild animals we found nearby, and that’s how we got through the pandemic.”

(LC16)

The quantitative analysis (Dimension 3: Deferral of Social and Environmental Costs) suggests that most respondents tend to downplay the negative impacts of tourism during its early stages. However, interviews reveal more severe criticisms, notably concerning visitor saturation and inadequate planning, particularly in communities such as Sayausí and Tsuer Entsa. In these areas, the rapid resurgence of tourism has exceeded local capacities for waste management and water regulation. Moreover, latent internal conflicts emerge—ranging from disputes over revenue control and accusations of environmental degradation to perceptions of inequitable benefit distribution—with some stakeholders explicitly opposing post-pandemic tourism, a sentiment not as distinctly captured in the quantitative data.

“The leaders lack self-management and focus solely on carrying out projects with the money they receive from tourism, but they have no vision for investment—only spending. There is no leadership in the community, which causes conflicts among association members. The board arbitrarily determines the salaries to be received. […] There is widespread corruption among the leaders, who expect to gain personal benefits from tourism. There is no transparency in management. They simply aim to receive a salary for two years (the duration of their term).”

(LC13)

While numerical data, particularly the PCA results, indicate moderate scepticism regarding state support, qualitative findings reveal a much more severe critique of the public sector—summarised bluntly as “just promotion, promotion, promotion”. The primary grievance among local stakeholders is not simply a lack of efficiency but a marked absence of risk management strategies and long-term planning. Moreover, the over-promotion of rural areas as “escape destinations” during the post-pandemic recovery has fostered unrealistic economic expectations and engendered conflicts over income distribution. This divergence between official narratives and community realities is far more pronounced in residents’ testimonies than in the statistical data.

“Within the public sector, we only have (…) a political figure (…) just promotion, promotion, promotion, and never any planning.”

(BS03)

“Promotion was never set aside. Now, the issue of promotions is very sensitive because promotion requires resources—creating more campaigns and figuring out how to proceed when the central government provided no funding and, therefore, no budget allocation for promotion. This meant there were no resources to ensure that international tourists would not forget about this wonderful destination waiting for them. While it is true that tourism was not possible during the pandemic […].”

(PS01)

Quantitative analysis (Dimension 2: Socioeconomic Participation) shows that some recognise the need to diversify their income sources. However, interviews reveal a more nuanced process. In communities such as Cluster Naranjal and Migüir, the necessity for diversification was not merely acknowledged but acted upon through initiatives like cacao micro-enterprises, community agriculture, and beekeeping. These measures, however, emerged primarily during the crisis and receded as tourism normalised. Thus, rather than a simple dichotomy between high and low engagement in non-tourism activities, actual behaviour fluctuated with economic conditions and the pressure to restore profitability in a sector long promoted by the state. This highlights a significant discrepancy between the “quantified intention” that diversification is beneficial and the observed practice, where tourism again took precedence during recovery.

“That was one of the main topics we discussed because it had already been decided that we would go into lockdown. So, I came up with the idea of launching an awareness campaign, which we managed to carry out in time: ‘We are here, don’t cancel your trip. We are still a destination. We will take this pause to prepare, but we remain a place you can visit.’”

(PS02)

The testimonies of interviewees illustrate that tourism continues to be perceived as a driver of immediate growth, with no clear structural changes aimed at fostering long-term resilience. Both the PCA and decision tree analysis indicate that most communities remain trapped in a cycle of “dependency” and “survival”, with limited adoption of transformative strategies. However, qualitative data provide a more nuanced perspective, revealing moments of genuine cooperation, such as joint adaptation initiatives and temporary partnerships within the private sector. These short-lived efforts, while not fully captured in quantitative analysis, demonstrate a latent resilience potential greater than what is reflected in numerical findings. Communities have shown the capacity to collaborate and manage resources more equitably, yet they lack the necessary support and strategic vision to sustain these collaborative networks over time. This situation reveals a paradox: while residents have become more critical of tourism, they continue to cling to a dependency-driven model, perpetuating their own vulnerabilities.

5. Discussion and Conclusions

The present study investigates whether residents’ attitudes towards tourism during the COVID-19 pandemic reflect resilience or mere survival in eight emerging tourism territories within the Cajas Massif Biosphere Area in southern Ecuador in case of a major crisis. We employed a mixed-methods research approach to analyse the responses of various stakeholders—residents, the private sector, and the public sector—to the crisis.

Overall, residents maintained positive attitudes towards tourism. However, these attitudes stem from what we term the “tourism mirage”, where only economic benefits are emphasised, while social and environmental costs are externalised. This phenomenon is particularly common in contexts with low tourism development, as less consolidated areas tend to exhibit a greater aspiration for tourism dependency, almost perceiving it as a panacea.

The private sector, in turn, recognises its reliance on self-initiative, particularly in times of crisis. Tourism stakeholders focused on mitigating immediate losses, adopting strategies that, while appearing resilient in the short term, were in reality driven by an urgent need for survival. Businesses continued to operate under business models validated in stable contexts, failing to anticipate the profound transformations required in response to the crisis.

Similarly, the public sector failed to diversify its strategies, continuing to promote tourism as if it were its sole responsibility. Operational and strategic changes were implemented without long-term planning, creating confusion between the necessary improvisation for survival and the adaptive capacity that defines resilience. In the short term, both responses may help mitigate the immediate impact. However, only resilience involves a deliberate process of transformation and organisational learning, strengthening the capacity to confront future adversities.

In the studied communities, tourism functioned contrary to expectations, serving as a complement rather than the backbone of the local economy. Less tourism-dependent territories demonstrated better performance during the crisis. The pandemic forced a diversification of productive activities in response to the widespread decline in tourism. While this diversification could be interpreted as positive, in many cases, it resembled a desperate survival response rather than a deliberate process of adaptation and resilience.

At the resident level, the confusion between resilience and survival stems from the difference between a reactive response and a proactive transformation. During the crisis, residents temporarily displayed values of reciprocity and solidarity. However, positive attitudes focused on short-term economic benefits obscured the acceptance of a tourism model that fails to drive structural changes within the community. In this context, decision-makers speculated in vain on tourism’s benefits as a means to demonstrate effective management, despite the lack of substantive transformations in governance or economic diversification.

From the health emergency, various actions were identified in the communities studied, manifested through temporary efforts of economic diversification, community reorganisation and adaptation to the new conditions imposed by the pandemic. However, in most cases, these responses lacked continuity over time, primarily due to the absence of sustained institutional support, limited structural planning, and internal tensions among local actors.

The Hat Museum, for example, resumed its training and promotional activities after the pandemic, although without significant transformation, remaining in a passive mode of institutional continuity. In the case of the Austro agroecological producers, commercialisation channels were strengthened through itinerant fairs and the establishment of collection centres. While this constituted a valuable tactical response, it did not evolve into a long-term strategy for productive transformation.

Baños, with an already consolidated local tourism base, resumed tourism activities with relative normality. However, the continued implementation of sanitary measures in thermal spas reflected an adaptive form of resilience focused on visitor safety. In contrast, Sayausí presented a more complex response: although the crisis motivated the formal creation of a tourism association, individual interests and internal conflicts led to the group’s fragmentation, resulting in disjointed and private initiatives that limited collective action.

Migüir stands out as one of the strongest examples of proactive resilience. Following the pandemic, the community successfully consolidated a tourism association, incorporated new stakeholders, signed agreements with local governments, and advanced strategies to improve road access, addressing one of the territory’s main structural weaknesses.

In the case of the 6 de Julio settlement, tourism activity shifted towards a production-oriented model focused on crab farming. A change in community leadership led to a detachment from tourism as a development strategy, maintaining a passive stance towards the sector. Meanwhile, the Shuar community of Tsuer-Entsa reverted to its pre-pandemic practises, with no clear evidence of institutional or community learning directed towards sustainable tourism development.

Lastly, the Naranjal cluster illustrates a more complex and dynamic process. After a phase of fragmentation and conflict between tourism stakeholders and opposing groups—which resulted in the forced closure of its waterfalls—some actors redirected their efforts towards agro productive activities and launched associative tourism ventures in new locations. This reveals a form of proactive resilience with the capacity for territorial reconfiguration.

While many of the responses were reactive or short-term, incipient forms of resilience with transformative potential emerged in certain contexts. The study thus confirms that true resilience is not limited to withstanding a specific crisis, but rather entails the ability to learn, adapt, and transform the structural conditions that underlie vulnerability.

Ultimately, genuine resilience would require communities to critically reassess and reinvent their relationship with tourism. This would involve diversifying productive activities and integrating cultural and environmental values into tourism development, ensuring that the adopted model strengthens both community identity and adaptive capacity in the face of future crises. Such a distinction is crucial for the formulation of tourism policies that aim to foster structurally robust and adaptable development over time.

The practical implications of this study suggest that tourism policies should promote strategies for diversification and structural transformation within communities. The COVID-19 pandemic demonstrated that communities could not depend solely on tourism. The communities studied are aware of this fact because tourism has been promoted as a panacea for less developed regions, and the reality rarely lives up to the expectations. This is especially the case when a crisis jeopardises a time-consuming process such as tourism development, which many expected to become their main livelihood. However, reality has caused tourism to become complementary to other productive activities, which is not perfect, but it is a magnificent opportunity to implement the framework for tourism as a supportive means for sustainable development, and of course for resilient communities.

On the other hand, in response to the fragmented and limited actions of the public sector during the health crisis, several measures could have been more effective in fostering resilience in the territories studied. Firstly, the implementation of programmes aimed at organisational strengthening and community conflict resolution would have helped to consolidate associations and collaborative networks, particularly in cases where local actors were willing to engage but lacked technical support. Likewise, the development of continuous training processes focused on tourism management, local governance, and digital tools could have enhanced local capacities to adapt to changing scenarios. In addition, economic support policies—such as recovery funds or incentives for productive diversification—would have allowed reactive responses to evolve into sustainable strategies. Lastly, improved coordination between levels of government and more structured planning could have enabled more coherent interventions, better adapted to the specificities of each territory, thus moving beyond the largely isolated or promotional actions observed during and after the pandemic.

Although academic literature insists that tourism can promote collective actions in the territories, the results of the current study clearly suggest that a dominant tourism perspective oriented towards economic growth fragments social structures. As a result, tourism management is marked by an individualistic business motivation, and the social base is not addressed (considered not relevant) in the communities. In other words, there are fractures between the various tourism stakeholders because there is no participatory governance system that promotes associativity and co-creation rather than pure competitiveness for individual growth.

Hence, tourism stakeholders returned to pre-pandemic normality, disregarding the fact that it was the social capital of the communities that kept them afloat during the pandemic. Indeed, our results show that communities have strengthened their collective actions to find livelihood mechanisms and resist the crisis. Unfortunately, this was not the case for the tourism sector. This situation undermined the potential of tourism as a development means, while communities realised that they might have overestimated such potential in the past.

Our findings differ from those of Qiu et al. [63], who emphasise the importance of government trust based on sound decision-making during times of crisis. However, the interviews revealed a significant gap between state provisions and effective support for communities, ultimately making resilience largely dependent on the residents themselves. While authorities at various levels focused their efforts on tourism promotion, they neglected fundamental aspects of infrastructure and superstructure, thereby perpetuating local vulnerability.

While Hall, Prayag, and Amore [69,72] argue that the long-term recovery of tourism systems requires coordination among various stakeholders, our findings indicate that such coordination was fleeting and failed to establish a sustainable restoration of tourism activity. This scenario supports the claims of Kamata [15] and Lamhour, Safaa, and Perkumienė [70], who suggest that organisations tend to prioritise immediate objectives at the expense of collective well-being. It also aligns with the perspective of Prayag [17], which posits that resilience emerges from interactions and processes across multiple scales, resulting in heterogeneous responses to adversity. In this sense, some communities exhibited resilient behaviours, but these were limited to the emergency context. Once the crisis subsided, individual interests resurfaced, suggesting a short-term survival strategy rather than a sustained, transformative resilience process.

As for methodology, we can confirm that mixed methods provide richer and more effective ways to study complex phenomena. While quantitative data revealed key variables in residents’ perceptions on tourism, qualitative data contributed to explain why the variables behave like that, to better understand their context and to make connections between them.

The findings of this study advance knowledge in the field of tourism by clarifying the distinction between resilience and survival in crisis contexts. They demonstrate how the responses of key stakeholders—residents, the private sector, and the public sector—vary depending on their level of dependence on tourism and their strategic approaches. Furthermore, by showing that tourism functions as a complement to the local economy in communities with low tourism development, this study offers an innovative perspective that challenges previous assumptions about the relationship between tourism and resilience in emerging destinations.

Moreover, our findings address key gaps identified in the literature review, particularly the lack of mixed-methods studies examining the implications of residents’ attitudes towards tourism in crisis situations such as the COVID-19 pandemic. This research provides empirical evidence on how diversification strategies can be interpreted either as survival responses or as manifestations of resilience, depending on the structural adaptive capacity of communities. These insights contribute to the formulation of tourism policies and management strategies that foster more sustainable and resilient long-term development.

This study demonstrates that the COVID-19 pandemic has acted as a catalyst, exposing the duality in the responses of tourism stakeholders. Communities, the private sector, and the public sector adopted strategies that appeared to indicate resilience, yet in many cases, these responses were driven by an immediate need for survival. The prevalence of the “tourism mirage”—where economic benefits are prioritised at the expense of social and environmental costs—highlights the inherent vulnerability of development models based on unilateral dependencies. Furthermore, the crisis underscored the insufficiency of management strategies that fail to consider productive diversification and the integration of cultural and environmental values—both of which are essential for strengthening communities’ adaptive capacity in the long term.

From a community resilience perspective, it is not enough to analyse the capacity of communities to adapt to disturbances such as the pandemic; it is also essential to consider their ability to reorganise, engage in collective learning, and transform structures that perpetuate vulnerability. In this sense, the development of resilience is closely linked to the principles of sustainable development, particularly with regard to strengthening social capital, fostering participatory governance, and ensuring balanced resource management. Our results are in line with literature indicating that for resilience to be effective, it must be inclusive, multisectoral, and long-term in orientation.

Moreover, the concept of the tourism mirage—used in this study to highlight the limits of unplanned tourism growth—requires further critical reflection. In the cases analysed, negative externalities were identified that disproportionately affect local communities, such as ecosystem degradation due to a lack of control in sensitive areas, the intensification of social conflicts over space use, and the emergence of unmet economic expectations. The absence of mitigation strategies—such as environmental management plans, mechanisms for equitable benefit distribution, or multi-stakeholder dialogue processes—has hindered the potential for tourism to serve as a genuine pathway for sustainable development.

Moreover, our findings challenge the dominant perspective that associates tourism exclusively with economic growth and social cohesion. Empirical evidence suggests that in low-tourism-development contexts, promoting tourism as a panacea can instead create divisions and fragment social structures, ultimately hindering genuine community resilience. This study not only fills a gap in the literature through the application of mixed-methods research, capturing the complexity of the phenomenon, but also provides critical insights for tourism policy formulation aimed at fostering structurally robust and adaptable development. Thus, this research underscores the urgent need to transform tourism models towards a participatory and collaborative governance approach, recognising the crucial role of social capital in building sustainable and resilient communities capable of withstanding future crises.

In short, this study clarifies the conceptual misunderstanding between tourism resilience and survival strategies, as revealed through stakeholder perceptions during the COVID-19 pandemic in Ecuador. Findings indicate that many local responses initially interpreted as evidence of resilience were predominantly survival-oriented measures aimed at securing short-term economic stability. Furthermore, government policies primarily focused on immediate tourism promotion rather than addressing underlying structural vulnerabilities, resulting in limited effectiveness and placing the burden of adaptation disproportionately on local communities.

Consequently, the primary contribution of this study lies in clearly distinguishing between genuine resilience and temporary survival responses within tourism contexts. It highlights the critical need for more comprehensive, inclusive, and long-term-oriented policy interventions that actively involve stakeholders, thereby strengthening social cohesion and enhancing the sector’s true adaptive capacity to withstand future crises. The findings emphasize that applying a one-size-fits-all policy is unsuitable due to the socio-spatial diversity observed across tourism contexts. This conceptual clarification offers both theoretical and practical implications for sustainable tourism development in Ecuador and comparable regions globally.

Limitations of the Study

Several limitations of this study should be noted. First, the research focuses on eight specific communities within the Cajas Massif Biosphere Area, a territory with particular socioeconomic, institutional, and territorial dynamics. As such, the findings are context-specific and their generalisability to other regions—especially those with different levels of tourism development, governance structures, or cultural frameworks—should be approached with caution. This said, the inclusion of a diverse range of cases within the study area enhances our understanding of how context-specific factors influence tourism dynamics—thus turning a potential limitation into a valuable comparative perspective.

While the study offers valuable insights into the interplay between tourism and community resilience in emerging destinations, it does not claim to provide universally applicable conclusions. Rather, it aims to contribute to a broader theoretical and methodological understanding of how rural and peri-urban communities respond to systemic crises such as the COVID-19 pandemic. Future research could apply and adapt the conceptual distinctions and analytical framework proposed here, to other socio-territorial contexts in order to assess their relevance, contrast findings, and refine policy recommendations across diverse tourism settings. Second, this study could have benefited from more quantitative indicators to objectively differentiate between survival strategies and resilience processes. This distinction was primarily addressed through qualitative analysis, which allowed for a more nuanced interpretation based on discourse and context. However, the development or application of measurable indicators could strengthen future research by offering a more systematic framework to distinguish between short-term adaptive responses and sustained, transformative resilience. Third, by focusing on a specific period during the pandemic, it is difficult to assess whether the identified behaviours—whether survival-driven or seemingly resilient—persisted over time or evolved after the most critical phase of the crisis had passed. Additionally, the use of surveys and interviews introduces the potential for response bias, particularly in a moment of health and economic instability. Moreover, this study primarily analyses residents’ attitudes without a deep exploration of the perspectives of indirect stakeholders (e.g., suppliers or civil society groups not directly involved in tourism). Finally, the perception of tourists is not considered, meaning that this study does not address how visitors experienced and understood the dynamics of these destinations during the crisis.