Destination Evaluation Attributes for Tourists in Hotel and Non-Hotel Accommodation in Spain

Abstract

Accommodation constitutes a key element in the competitiveness of a destination. However, there is limited information on how overnight tourists evaluate the destination according to the type of accommodation. To date, Information and Communication Technologies (ICTs) have led to an increased use of online media and the generation of large datasets available for analysis. This study analyses 47,568 feedback surveys administered by Spanish accommodation establishments to tourists, provided anonymously by a tourism software company. The main objective is to understand which destination attributes predict positive results on satisfaction and recommendation, depending on the type of accommodation (hotel and non-hotel). To this end, both a descriptive analysis and an analysis using neural networks are conducted. The results reveal significant differences in the evaluation of destination elements depending on whether the accommodation is a hotel or non-hotel, with the predictive variables varying across each typology. As a key conclusion, the study highlights that adopting this perspective makes it possible to understand destination evaluation from the standpoint of overnight tourists, thereby contributing to destination-related literature. From a practical perspective, recommendations are offered to both destination managers and accommodation providers.

1. Introduction

Tourism is a relevant sector in the economic development of a territory, generating value in terms of local economy, wealth growth and improvement of public services (Barrera-Martínez et al., 2025). There is a high degree of competitiveness between tourist destinations, but various indicators can reflect the effectiveness and leadership of some destinations over others. Spain is one of the main tourist destinations in Europe in terms of arrivals and nights spent. As Eurostat Statistics indicate, Spain is the fourth European country in terms of nights spent, with a total of 183,129,528 in 2023 (Eurostat, 2023b); and the third European destination in tourist arrivals (72,690,996 in 2023) (Eurostat, 2023a). This data highlights the relevance of overnight stay tourism in this country. In the tourism industry, there are some differences based on the tourist motivations, stay and behaviour. Being all of them visitors (UN Tourism, 2025), to unify the terms for evaluating statistics and data, defines two kinds of them. On one hand, the tourist or overnight visitor is classified as a visitor who spends at least one night in the destination. Alternatively, the excursionists or same-day visitors are defined as visitors who do not stay in the destination more than 24 h, without contracting any accommodation service. These differences are also related to the behavioural and satisfaction results and expectations of the trip (Xu et al., 2018).

The tendency of tourists to stay in Spanish destinations is highly related to the accommodation development. In tourism, accommodation development is one of the main value-created tools that improves the tourist expenditure but also affects the destination competitiveness (Barrera-Martínez et al., 2025). In this sense, there is no unique type of accommodation and the development of them is different as accommodation types influence the capacity of the destination’s tourism system (Barrera-Martínez et al., 2025). Subsequently, each country regulates and defines various accommodation classifications. In general, hotel and non-hotel accommodations are related to two different supplies, including diverse markets, needs and services.

The study of the variables that may influence the evaluation of overall satisfaction is of particular importance, as no conclusive studies have determined which factors, in general, influence global satisfaction (Travar et al., 2022). The evaluative elements of the destination have been defined in various ways and approached from different perspectives. In terms of terminology, they can be described as attributes (Braimah et al., 2024), indicators, factors (Alegre & Garau, 2010), or determinants (Omo-Obas & Anning-Dorson, 2023). Jumanazarov et al. (2020) argue that no standardised scale of destination attributes exists, and that the number of elements may vary depending on their relevance to a specific study. Recent research has shown that attributes represent parameters or individual assessments made by tourists regarding different aspects of a destination. Eusébio and Vieira (2013) note that the literature contains a significant number of studies that distinguish between tourists’ evaluation of destination attributes and their overall satisfaction.

The above factors serve different purposes depending on the stage of the tourism experience. After the trip, a positive evaluation of the destination by tourists may contribute to the formation of the destination image (Badoni et al., 2025), perceived quality (Cheraghzadeh et al., 2024), or competitiveness (Amaya-Molinar et al., 2017). Conversely, before the trip, such elements may influence destination choice (Badoni et al., 2025; Mussalam & Tajeddini, 2016) or act as motivational drivers. By comparing their prior expectations with the factors experienced during their stay at the destination, tourists construct their level of satisfaction (Braimah et al., 2024).

Moreover, the tourism sector is currently immersed in a context of digitalisation in which tourists engage in an online post-travel stage that facilitates the transmission of feedback to service providers, generating valuable data for analysis. An example of this is feedback surveys (Fuchs et al., 2014). To examine whether there are differences in destination evaluation by tourists depending on the type of accommodation chosen, the use of big data becomes highly relevant, as it allows for a deeper understanding of tourist demand, satisfaction, and behaviour (Li et al., 2018). Furthermore, statistical analyses represent a substantial proportion of big data analysis techniques. In this regard, the present study draws on feedback surveys collected from Spanish accommodation establishments.

Taking all the above into account, the present study considers how destinations are evaluated by tourists who stay at them. The main objective is to understand which destination attributes are considered for satisfaction and recommendation depending on the type of accommodation. This main objective includes two specific objectives. The first is focused on determining the influence of the type of accommodation establishment (i.e., hotel and non-hotel) on the evaluation of destination elements, satisfaction, and recommendation intentions. The second seeks to identify the extent to which destination elements function as predictors of visitors’ satisfaction and recommendation intentions. To achieve these aims, different methodological tools are employed.

2. Theoretical Framework

2.1. Overnight Tourists

Destination management organisations’ (DMOs) main aim is to increase the expenditure and the length of stays because it is translated into benefits for governments, businesses and the local population (Xu et al., 2018; Oh & Schuett, 2010). Additionally, overnight stays in a tourist destination are proven to: (i) improve the tourists’ enjoyment and expenditure, (ii) reduce some negative externalities, (iii) motivate the destination exploration (Yuksel et al., 2010; Xu et al., 2018). Also, overnight tourists who spend at least one night at the destination have particular characteristics and behaviours. Some of the reasons that motivate the increase in stays are related to the tourists’ personality, lifestyle and attitudes towards seeking new experiences (Kim et al., 2022).

Different studies certainly highlight that sleep abroad allows: (i) in terms of tourists, to improve the satisfaction and the experience through the relaxation and through the possibility of getting to know the city; (ii) in terms of destination, can increase tourist income and contracting complementary services (Xu et al., 2018). These implications mean that both destination administrations and the business network must ensure the quality of the tourism experience in an integrated manner (Monsalve-Castro & Hernández-Rueda, 2015).

Some of the tourists’ characteristics that can lead to differences in their destination behaviour are age, gender, trip motivation, group type, budget, tourism typology or accommodation service selection, based on the Equity Theory by Oliver and Swan (1989). Among the aforementioned, some authors suggest the relevance of accommodation typology as an effective segmentation criterion for analysing the tourist behaviour (McKercher et al., 2023).

2.2. Hotel and Non-Hotel Accommodation

Accommodation is one of the main attributes in the destination management and selection (Radojevic et al., 2018; Rašovská et al., 2021; Lin et al., 2024). Destination tourism research has been focused on how accommodation is chosen, what typologies are better for the development in diverse destinations or how these services impact the overall satisfaction, recommendation or some other marketing results such as loyalty or competitiveness (Radojevic et al., 2018; Kim et al., 2022; Lin et al., 2024; Barrera-Martínez et al., 2025). But there is not enough information about how tourists, who spend nights in an accommodation service, evaluate the destination (Rašovská et al., 2021). There are differences in the tourists’ behaviours in terms of accommodation typologies. Some research highlights that the accommodation typology selection can predict the tourists’ behaviours (Rašovská et al., 2021). For instance, hotel services are preferred by elderly tourists who spend more money (Pestana et al., 2020), but non-hotel accommodations have increased in terms of tourists’ stays during the last years (Gonzalez-Diaz et al., 2015).

Specifically, it has been proven that hotel experience contributes to a better overall experience (Pestana et al., 2020) because destination resources also motivate the pre-selection of this type of accommodation (Radojevic et al., 2018). This is why hotel services contribute to the positioning and promotion of the destination, leading to greater tourist satisfaction through quality (Monsalve-Castro & Hernández-Rueda, 2015).

On the other hand, different studies have assessed the impact of non-hotel tourism services. Some of the conclusions are related to the idea that non-hotel accommodations are key elements of destination image and experience (Lalicic et al., 2021), and that this type of accommodation leads to the development of more destination activities and time spent and, therefore, to more enjoyment and satisfaction (McKercher et al., 2023; Tussyadiah & Pesonen, 2016) In this regard, studying the differences between the category of accommodation establishments has important implications for destination management and marketing (Barrera-Martínez et al., 2025; McKercher et al., 2023).

2.3. Destination Evaluation Attributes

Braimah et al. (2024), drawing on Dann (1981), establish that destination elements can be classified as push factors, which correspond to the tourist’s internal needs that drive individuals to travel, and pull factors, which refer to the destination’s inherent characteristics that motivate tourists to choose a specific destination once the decision to travel has been made.

In this sense, Omo-Obas and Anning-Dorson (2023) identify cognitive factors as those related to the tourist’s experience or perceived image, also including within this category the motivational factors, divided into push and pull. As a consequence of the assessment of these factors, affective factors emerge, such as satisfaction, trust, and destination involvement.

Moreover, as previously noted, destination evaluation attributes contribute directly to tourist satisfaction. Atsız and Akova (2021) indicate that destinations possess distinctive attributes that, in themselves, attract tourists, who then evaluate them to determine their satisfaction. Alegre and Garau (2010) argue that motivational factors can lead to both satisfaction and dissatisfaction with the destination. Hence, there are genuinely motivational factors that generate positive satisfaction responses, whereas others, of a hygienic nature, are associated with negative responses. Notably, numerous studies have adopted this perspective when addressing destination satisfaction.

According to Braimah et al. (2024), the interaction between push and pull factors gives rise to destination satisfaction, with the evaluation of destination-specific attributes being particularly decisive. For this reason, in this study, a distinction is made between variables linked to specific destination attributes and those related to evaluations more closely associated with marketing outcomes, such as satisfaction with destination services, evaluation of the area, or intention to recommend.

With regard to the first classification, Alegre and Garau (2010) identify a set of 24 general destination evaluation attributes, including cultural or sports activities, information, tourist attractions, accommodation, and gastronomy. Badoni et al. (2025) also incorporate factors associated with the destination’s tourism resources, such as infrastructure, transportation, and value-added services. Similarly, Amaya-Molinar et al. (2017) highlight twelve key factors, including variables linked to destination marketing, such as infrastructure, transportation, management, and marketing itself. In addition, Cheraghzadeh et al. (2024) include in their analysis factors related to overall destination evaluation, such as food quality, infrastructure, and accessibility.

Furthermore, other classifications are also found, such as that proposed by Vojtko et al. (2022), who distinguish between endogenous factors, controllable by destination managers (e.g., safety, hospitality, friendliness of locals, cleanliness, transport infrastructure, or visitor management level), and exogenous factors, uncontrollable, such as weather conditions or climate in general. In the same vein, Mutinda and Mayaka (2012) differentiate between environmental factors, including recreational activities, available information, ease of access, and outdoor activities. Since activities are often broken down by their characteristics and typology, their evaluation may be considered both in terms of quantity and adequacy. Additionally, Mutinda and Mayaka (2012) identify a second category of factors: individual trait factors, which they associate with push factors.

From a different perspective, Fallon and Schofield (2006) classify these factors according to their importance and performance into four groups: excitement factors, important performance factors, unimportant performance factors, and basic factors. For first-time visitors, restaurant evaluation and information are considered basic, while accommodation is deemed important. For repeat visitors, the factors remain similar; however, customer service and the price–quality ratio of restaurants are reclassified as important performance factors. Thus, regarding restaurant services, these can be evaluated both by their quality and by their quantity.

Taken together, it can be determined that destination attributes are essential in shaping the overall evaluation of a destination. Nonetheless, as previously noted, differences may exist depending on the type of accommodation. Consequently, the following hypothesis is proposed:

H1. 

The evaluation of destination attributes differs according to the category of accommodation establishment (i.e., hotel and non-hotel).

In the existing scientific literature, overall satisfaction and intention to recommend are commonly considered outcome variables. In this regard, several authors have noted that multiple endogenous and exogenous factors, both controllable and uncontrollable, can influence tourists’ overall satisfaction (Vojtko et al., 2022; Štumpf et al., 2022). Among them, destination quality—comprising elements such as accessibility, accommodation, and tourist attractions—has been identified as a key determinant of satisfaction (Travar et al., 2022). Likewise, Štumpf et al. (2022) highlight the evaluation of services, transportation, and tourism supply as relevant factors, also underlining the role of accommodation as a decisive variable in the overall experience.

Complementarily, Jumanazarov et al. (2020) conclude that destination attributes influence visitors’ cognitive and affective evaluations, which, in turn, condition satisfaction. Satisfaction is closely linked to tourists’ future intentions, particularly conative dimensions such as recommendation and word-of-mouth communication, which directly affect destination image (Travar et al., 2022). Within this framework, accommodation satisfaction emerges as a factor that can foster both revisitation and destination recommendation (Schofield et al., 2020). Moreover, Romão et al. (2014) incorporate the number of overnight stays as a significant indicator in the assessment of overall satisfaction.

Regarding the relationship between satisfaction and recommendation intentions, various studies consistently point to a positive correlation (Jumanazarov et al., 2020; Papadopoulou et al., 2023) although some works have not been able to corroborate this association significantly (Eusébio & Vieira, 2013).

In summary, and given the importance of these concepts, the following hypothesis is proposed to examine whether differences exist between these two outcome variables depending on the type of accommodation. Therefore, the following hypotheses is proposed:

H2. 

The evaluation of satisfaction and recommendation intentions differ according to the category of accommodation establishment (i.e., hotel and non-hotel).

H3. 

The destination attributes have different levels of importance when predicting favourable levels of satisfaction and behavioural intentions based on the category of accommodation establishment (i.e., hotel and non-hotel).

The main theoretical contribution of this study lies in addressing a gap identified in the existing literature. Previous research has examined destination satisfaction attributes, often providing general lists without classifying them into specific analytical factors (Alegre & Garau, 2010; Štumpf et al., 2022). Moreover, these attributes have rarely been employed as the primary object of analysis. In most cases, they are used merely as instrumental variables within factorial or structural models in which satisfaction (Zulvianti et al., 2022), revisit intention (Jumanazarov et al., 2020; Eusébio & Vieira, 2013; Braimah et al., 2024), or loyalty (Omo-Obas & Anning-Dorson, 2023) constitute the ultimate outcome variables. Consequently, previous studies have not focused on identifying which attributes play a more significant role in shaping cognitive or affective evaluations; rather, they have been limited to determining whether such relationships exist. Mussalam and Tajeddini (2016) explored destination attributes in greater depth, but their analysis primarily examined how these attributes influence evaluation and choice.

In contrast, the present research investigates the behaviour of destination attributes in relation to both satisfaction and recommendation, considering the distinction between hotel and non-hotel accommodation, as Figure 1 shows. Furthermore, it introduces a new classification of destination attributes derived from an extensive literature review, as presented in

Table 1. Survey questions analysed.

Item Code	Construct	Definition
[DEST_1]	Tourism resources of the area (Alegre & Garau, 2010; Eusébio & Vieira, 2013; Mussalam & Tajeddini, 2016; Travar et al., 2022; Štumpf et al., 2022)	All natural or cultural elements available in a given destination.
[DEST_2]	Adequacy of tourism resources (Fallon & Schofield, 2006; Jumanazarov et al., 2020; Eusébio & Vieira, 2013; Celotto et al., 2015; Mussalam & Tajeddini, 2016; Schofield et al., 2020; Omo-Obas & Anning-Dorson, 2023)	Implementation of infrastructures, facilities, equipment, accessibility elements, and signage that enable their use by visitors.
[DEST_3]	Possibility of engaging in activities (Mutinda & Mayaka, 2012; Jumanazarov et al., 2020; Schofield et al., 2020; Travar et al., 2022; Štumpf et al., 2022)	The extent to which tourists perceive the existence and accessibility of tourism-related experiences and activities at the destination.
[DEST_4]	Tourism information services (Mussalam & Tajeddini, 2016; Romão et al., 2014; Jumanazarov et al., 2020; Štumpf et al., 2022)	Set of resources and tools designed to assist, inform, and provide guidance to tourists regarding the destination’s offerings.
[DEST_5]	Quantity of food service provision (Fallon & Schofield, 2006; Celotto et al., 2015; Jumanazarov et al., 2020; Wu et al., 2022)	The number of establishments and facilities providing gastronomic and beverage offerings in the destination.
[DEST_6]	Quality of food service provision (Eusébio & Vieira, 2013; Mussalam & Tajeddini, 2016; Jumanazarov et al., 2020; Schofield et al., 2020; Travar et al., 2022; Zulvianti et al., 2022)	The perceived quality of establishments and facilities offering gastronomy and beverages at the destination.
[DEST_7]	Overall assessment of services (Eusébio & Vieira, 2013; Jumanazarov et al., 2020; Atsız & Akova, 2021; Štumpf et al., 2022; Papadopoulou et al., 2023; Omo-Obas & Anning-Dorson, 2023; Braimah et al., 2024)	Visitor’s assessment by comparing prior expectations with the experience at the destination, integrating both rational evaluations of attributes and emotional responses (Lalicic et al., 2021).
[DEST_8]	Overall assessment of the area (Eusébio & Vieira, 2013; Mussalam & Tajeddini, 2016; Schofield et al., 2020; Papadopoulou et al., 2023; Badoni et al., 2025)	The general assessment of the destination, considering its overall offer and evaluation of attributes.
[DEST_9]	Willingness to recommend to a friend (Eusébio & Vieira, 2013; Mussalam & Tajeddini, 2016; Jumanazarov et al., 2020; Sun et al., 2020; Papadopoulou et al., 2023)	The traveller’s intention to recommend the destination to friends, relatives, and other potential tourists.

Source: Own elaboration.

.

3. Materials and Methods

3.1. Data Collection

This study is based on an exploratory approach and employs a quantitative methodology. The data collected correspond to surveys administered to tourists in Spanish accommodation establishments. For the purposes of this study, the sample has been divided according to the type of accommodation establishment, specifically into hotel and non-hotel categories. According to McKercher et al. (2023), analysing tourist perceptions by accommodation type is relevant, since each customer exhibits different behaviour depending on the chosen lodging option.

The data derived from tourists generates two types of feedback or knowledge: explicit and implicit, depending on whether it is provided intentionally or not. The former refers to data intentionally supplied, such as surveys collected by different accommodation companies, while the latter refers to data unintentionally generated, such as booking behaviour (Höpken et al., 2018).

This study evaluates the post-travel experience of tourists at the destination and whether differences exist in its assessment according to the accommodation category in Spanish establishments, based on feedback surveys administered by the accommodations to their clients. The aim is to gain deeper insights into the tourist experience with the destination through these surveys.

The feedback surveys analysed correspond to the period June 2022 to June 2023, as these dates coincided with the post-COVID recovery of tourism businesses worldwide (Ku & Chun-Der, 2024), and particularly in Spain. The dataset was provided by the company MisterPlan (www.misterplan.es), comprising 47,568 anonymised feedback surveys from both hotel and non-hotel accommodations. The initial dataset comprised 50,790 anonymized post-stay feedback surveys. During data cleaning, 3222 incomplete responses, either blank or containing more than three missing variables were removed. The final sample used for analysis therefore consisted of 47,568 valid observations. The survey was distributed via email to the person responsible for the booking, shortly after the end of the stay. As such, only one response per booking could be received, ensuring consistency in the unit of analysis.

The questions included in the survey are presented in Table 1. The items DEST1, DEST2, DEST3, DEST4, DEST5, and DEST6 correspond to destination attributes. These items were measured using a five-point Likert scale, where 1 = “not important at all” and 5 = “very important.” Respondents were asked to rate the importance of each attribute based on their experience during the visit. In addition, items DEST7 and DEST8 represent marketing results reflected by the evaluation of tourists’ satisfaction, while item DEST9 measures their behavioural intentions. All three items were assessed using a five-point Likert scale, where 1 = “strongly disagree” and 5 = “strongly agree.” Participants were asked to indicate the extent to which they agreed or disagreed with each statement. These nine indicators correspond to the exact wording of the items seen by respondents. The survey was originally administered in Spanish; however, for the purposes of this study, it was translated into English. The translation was subsequently validated by expert researchers in the field with a high proficiency in English to ensure that no errors altered the meaning of the items. Finally, the questionnaire consists of single-item measures, which are considered a suitable approach in exploratory research, as is the case in this study (Diamantopoulos et al., 2012).

3.2. Research Scenario

Participating establishments include both hotel and non-hotel accommodation (such as rural houses, tourist apartments, and campsites) across multiple Spanish regions. While MisterPlan clients may not fully represent the national accommodation mix, they cover a diverse range of destinations and business types, which contributes to the heterogeneity of the sample. The response rate could not be precisely determined, as the total number of survey invitations distributed was not reported by the provider. However, the large number of valid responses (n = 47,568) provides strong analytical robustness. To minimise non-response bias, the survey was anonymous and standardised across all participating establishments, encouraging participation.

The distribution and relevance of hotel and non-hotel accommodation in Spain have been previously evaluated as a subject of great interest in the Academy due to its economic and social repercussions in the territories and at the national level (Alegre & Garau, 2010; International Business Machines, 2019; Pulido-Fernández et al., 2024). In Spain, there are differences between hotel and non-hotel categories, but both imply relevance in the creation of tourism value in the Spanish destination. Firstly, in terms of the number of establishments and capacity, the hotel and non-hotel typologies represent a large tourism business force (

Table 2. Number and bed places of accommodation by typology.

Accommodation Typology		Number	Bedplaces
Hotel accommodation (rural and urban)		12,798	1,163,339
Non-hotel accommodation	Camp sites	522	272,705
	Hostels	1203	69,269
	Tourist apartments	6178	421,474
	Rural tourism accommodations	15,482	140,359
	All typologies	23,363	903,807

Source: Data obtained from the INE with the latest data for the year 2023 (INE, 2025b, 2025a).

). But also, in terms of the length of stay. Hotel accommodations result in 3.09 nights as an average stay in 2023, obtaining higher rates in international tourists (3.92) than national ones (2.22) (INE, 2025b). Instead, some other non-hotel accommodations, including apartments, rural services and active tourism accommodations, among others, have an average stay of around 1.5 and 3.17, being of great interest the accommodations that include extra services (apartments) and related to rurality (rural tourism accommodation) (INE, 2025a).

3.3. Data Analysis

For data analysis, from the total of the surveys administered (n = 47,568), 23,943 belong to the hotel category, and 23,625 to the non-hotel. For statistical analysis, both the subsets corresponding to each accommodation category were analysed. First, a descriptive approach was applied to determine the importance attributed by tourists to each of the indicators analysed. Subsequently, to verify the research hypotheses, a t-test for independent samples was conducted to determine the influence of accommodation category (i.e., hotel and non-hotel) on the evaluation of destination attributes, satisfaction, and recommendation intentions. The t-test for independent samples is one of the most widely used parametric tests for examining differences between two distinct groups (Verma & Salam, 2019). Furthermore, the t-test has been shown to be robust across a wide range of conditions when sample sizes are large (Rochon et al., 2012).

Finally, to corroborate the role of destination attributes to predict marketing outcomes a neural network analysis was applied. Escandón-Barbosa and Hurtado-Ayala (2014) indicates that neural networks group elements based on certain variables, obtaining the level of importance of these variables in influencing business results. To do this, artificial neural networks are information processing systems in which data is processed through elements called nodes or neurons that are organised in layers. Each node or neuron is connected to other neurons by communication links, which have an associated weight, allowing the neural network to recognise a specific problem to be analysed. Two types of neural network models are identified: multilayer perceptron and radial basis function. For this study, the multilayer perceptron model will be employed, as it generates a predictive model for one or more dependent variables based on the values of the predictor variables. For the descriptive analysis, the t-test for independent samples, and the neural networks the IBM SPSS 29.0.0.0.0 statistical programme was applied. The results obtained are reported in the following section.

4. Results and Discussion

4.1. Assessment of Destination Attributes

Table 3. Assessment of the importance of destination attributes.

Destination Attributes	1	2	3	4	5	M
Tourism resources of the area	0.5	1.0	6.6	28.8	63.2	4.53
Adequacy of tourism resources	5.0	1.2	8.4	33.2	56.8	4.45
Possibility of engaging in activities	6.0	1.4	9.2	29.9	59.0	4.45
Tourism information services	1.1	2.5	13.6	31.1	51.7	4.30
Quantity of food service provision	1.5	2.8	11.7	29.4	54.7	4.33
Quality of food service provision	1.3	2.1	10.5	30.8	55.4	4.37

Note. Values for response categories (1–5) are expressed as percentages; M = Mean. Source: Own elaboration.

summarises the results of the analysis of the destination attributes for the entire sample (n = 47,568). Based on the scores obtained, most tourists assigned a degree of importance between 4 (i.e., important) and 5 (i.e., very important) to all indicators. However, when analysing the average score, the importance of three elements stands out: ‘the tourism resources of the area’ (4.53), ‘the adequacy of tourism resources’ (4.45) and ‘the possibility of engaging in activities’ (4.45). This indicates that the positive assessment of a destination depends to a greater extent on the tourist resources it possesses, its adaptation to facilitate the tourists’ experience, and the variety of activities available.

Table 4. Assessment in terms of marketing results.

Marketing Results	1	2	3	4	5	M
Overall assessment of services	0.6	1.3	7.5	33.2	57.4	4.46
Overall assessment of the area	0.4	0.9	5.7	29.6	63.4	4.55
Willingness to recommend to a friend	1.1	1.1	4.0	21.9	71.8	4.62

Note. Values for response categories (1–5) are expressed as percentages; M = Mean. Source: Own elaboration.

reflects the results of tourists’ assessment of the services, the place visited, and their willingness to recommend the destination to a friend. The findings indicate that, although the scores achieved have been quite favourable, tourists show a higher level of satisfaction with the place visited (4.55) rather than with the services received (4.46). In addition, there is a high willingness to recommend the destination to others (4.62). Overall, these results suggest that tourists have had a positive experience at the destination.

4.2. Assessment of Destination Attributes by Type of Accommodation

4.2.1. Hotel Category

The results reported in

Table 5. Assessment of the importance of destination attributes based on the perceptions of tourists staying in establishments within the hotel category.

Destination Attributes	1	2	3	4	5	M
Tourism resources of the area	0.5	1.2	7.1	29.6	61.5	4.50
Adequacy of tourism resources	0.5	1.4	9.4	34.0	54.7	4.41
Possibility of engaging in activities	0.6	1.5	10.4	31.2	56.3	4.41
Tourism information services	1.1	2.7	14.5	32.0	49.7	4.26
Quantity of food service provision	1.4	2.8	11.3	29.8	54.7	4.34
Quality of food service provision	1.2	2.2	10.6	31.2	54.7	4.36

Note. Values for response categories (1–5) are expressed as percentages; M = Mean. Source: Own elaboration.

summarise the evaluation of the destination attributes carried out by tourists staying in establishments within the hotel category. Overall, respondents assigned high importance to all indicators, with mean scores ranging between 4 (i.e., important) and 5 (i.e., very important). Analysis of the average scores indicates that ‘tourism resources of the area’ (4.5), ‘Adequacy of tourism resources’ (4.41) and ‘possibility of engaging in activities’ (4.41) were perceived as the most important aspects. In contrast, although still highly rated, ‘restaurant quality’ (4.36), ‘restaurant quantity’ (4.34) and ‘tourist information services’ (4.26) were considered slightly less important.

The findings presented in

Table 6. Assessment of marketing results based on the perceptions of tourists staying in hotel category establishments.

Marketing Results	1	2	3	4	5	M
Overall assessment of services	0.6	1.4	8.0	34.1	55.9	4.43
Overall assessment of the area	0.4	1.0	6.6	31.0	61.1	4.51
Willingness to recommend to a friend	1.2	1.2	4.8	23.6	69.3	4.59

Note. Values for response categories (1–5) are expressed as percentages; M = Mean. Source: Own elaboration.

detail tourists’ perceptions of destination services, the place visited, and their willingness to recommend the destination. Among tourists staying in hotel establishments, the analysis reveals a slightly higher level of satisfaction with the place visited (4.51) compared to the services received (4.43); although both scores indicate generally positive evaluations. In addition, tourists expressed a strong intention to recommend the destination (4.59), reinforcing the favourable overall perception of their stay.

4.2.2. Non-Hotel Category

As shown in

Table 7. Assessment of the importance of destination attributes based on the perceptions of tourists staying in non-hotel category establishments.

Destination Attributes	1	2	3	4	5	M
Tourism resources of the area	0.4	0.8	6.0	28.0	64.8	4.56
Adequacy of tourism resources	0.4	1.0	7.4	32.3	58.9	4.48
Possibility of engaging in activities	0.5	1.2	8.1	28.5	61.7	4.50
Tourism information services	1.1	2.3	12.7	30.2	53.7	4.33
Quantity of food service provision	1.5	2.9	12.0	28.9	54.6	4.32
Quality of food service provision	1.4	1.9	10.4	30.3	56.0	4.38

Note. Values for response categories (1–5) are expressed as percentages; M = Mean. Source: Own elaboration.

, most tourists staying in non-hotel establishments are assigned high importance ratings between 4 (i.e., important) and 5 (i.e., very important). The analysis of the average scores indicates that ‘the tourism resources of the area’ (4.5), ‘adequacy of tourism resources’ (4.41) and ‘possibility of engaging in activities’ (4.41) were perceived as the most relevant attributes for visitors. This coincides with the assessment made by tourists staying in the hotel category establishments. These results mirror those obtained for tourists staying in hotel establishments, suggesting that, regardless of accommodation type, the availability and suitability of local tourism resources, together with the diversity of activities offered, are key determinants of a positive destination evaluation.

Table 8. Assessment of marketing results based on the perceptions of tourists staying in non-hotel category establishments.

Marketing Results	1	2	3	4	5	M
Overall assessment of services	0.6	1.2	7.0	32.2	59.0	4.48
Overall assessment of the area	0.3	0.7	4.9	28.2	65.8	4.58
Willingness to recommend to a friend	1.0	1.0	3.3	20.3	74.4	4.66

Note. Values for response categories (1–5) are expressed as percentages; M = Mean. Source: Own elaboration.

shows the assessment about services, the place visited, and their willingness to recommend the destination. The analysis shows that tourists reported slightly higher satisfaction with the place visited (4.58) than with the services received (4.48). In addition, their intention to recommend the destination was high (4.66), reflecting generally positive experiences among visitors. These findings coincide with the perceptions of tourists staying in the hotel category establishments. Therefore, in both cases, it can be deduced that visitor satisfaction is determined to a greater extent by the experience with the place visited.

4.3. Independent Samples t-Test: Hotel Accommodation vs. Non-Hotel Accommodation

In this section, an independent sample t-test was performed to determine the influence of the type of accommodation (i.e., hotel and non-hotel) on the evaluation of destination attributes, satisfaction, and recommendation intentions. The results presented in

Table 9. Independent samples t-test.

	Hotel Category		Non-Hotel Category		Confidence Interval	p Value
Items	Mean	SD	Mean	SD
Destination attributes
DEST_1	4.50	0.729	4.56	0.679	[−0.070; −0.045]	<0.001 ***
DEST_2	4.41	0.759	4.48	0.711	[−0.087; −0.060]	<0.001 ***
DEST_3	4.41	0.785	4.5	0.740	[−0.100; −0.072]	<0.001 ***
DEST_4	4.26	0.885	4.33	0.865	[−0.082; −0.050]	<0.001 ***
DEST_5	4.34	0.885	4.32	0.902	[−0.003; 0.030]	0.111 ns
DEST_6	4.36	0.848	4.38	0.850	[−0.031; 0.000]	0.056 ns
Marketing results
SAT_1	4.43	0.750	4.48	0.728	[−0.060; −0.034]	<0.001 ***
SAT_2	4.51	0.698	4.58	0.653	[−0.083; −0.058]	<0.001 ***
REC_1	4.59	0.741	4.66	0.685	[−0.088; −0.063]	<0.001 ***

Note. Equal variances were assumed for all items, except for DEST 4 and DEST 5; SD = Standard deviation; Significance *** p < 0.001; ns = not significant. Source: Own elaboration.

indicate that, except DEST_5 and DEST_6, significant differences were found between the perceptions of tourists staying in hotel category establishments and those staying in non-hotel establishments.

In addition to assessing the statistical significance of group differences, it is recommended to estimate the magnitude of these differences by evaluating effect sizes (d-value). Values of 0.2, 0.5, and 0.8 correspond to small, medium, and large effects, respectively (Bakker & Wicherts, 2014). The observed effect sizes for items showing significant differences were: DEST1 (d = 0.704), DEST2 (d = 0.735), DEST3 (d = 0.763), DEST4 (d = 0.875), SAT1 (d = 0.739), SAT2 (d = 0.676), and REC1 (d = 0.714), indicating medium to large effect sizes for the differences between groups in all cases.

The results presented above support hypothesis 1 and 2, the assessment of destination attributes differ between accommodation categories. Several authors argue that the choice of accommodation affects, in some way, the evaluation of the destination (McKercher et al., 2023). Authors such as Barrera-Martínez et al. (2025) assert that the type of accommodation influences the tourism system, promoting elements of the local economy. This information finds empirical support in the data from this study. Furthermore, these findings indicate that the category of accommodation establishment significantly influences the way tourists assess the elements that contribute to their experience at the destination, and on the assessment of the experience itself in terms of satisfaction and recommendation intentions. This may be explained by Equity Theory, which posits that different variables, such as the type of stay, duration, type and motivation, affect satisfaction through attachment to the destination (Oliver & Swan, 1989).

4.4. Neural Network Analysis

The following sections present the results of the neural network analysis carried out both globally and distinguishing between accommodation categories (i.e., hotel and non-hotel). This analysis aims to identify the extent to which the destination attributes predict visitor satisfaction and recommendation intentions. To this end, six input or independent variables were included in the model design: tourism resources of the area, adequacy of tourism resources, possibility of engaging in activities, tourism information services, quantity of food services, and quality of the food service provided. The output or dependent variables comprised: overall assessment of services, overall assessment of the area and willingness to recommend to a friend. It should be noted that neural network analysis identifies associative rather than causal relationships among variables. Therefore, the results reflect statistical associations between the independent and outcome variables.

For the training phase of the neural network model, the data partitions were specified as follows: 70% for the training sample and 30% for the test sample. The hyperbolic tangent was employed as the activation function for the hidden layer, and the identity function was used for the output layer. To select the optimal model, several neural networks were trained by varying the number of neurons in the hidden layer. The training stopped if the error on the validation set did not improve after one consecutive step. These parameters were applied to the construction of the neural networks. However, the number of layers and neurons is specific to each model, as will be explained later.

In addition, an independent variable importance analysis was conducted. This procedure involves a sensitivity analysis that quantifies the relative contribution of each predictor variable to the performance of the neural network model. The analysis was carried out using the combined training and testing samples. The resulting output includes a chart that illustrates the normalised importance values for each predictor. Normalised importance is calculated by dividing each importance value by the highest observed value and expressing the result as a percentage. The importance chart displays the predictors in descending order of their relative importance (International Business Machines, 2019). The following sections present the results of the importance analysis conducted both globally and distinguishing between hotel and non-hotel categories.

4.4.1. Neural Network Analysis at a Global Level

Figure 2 illustrates the structure of the constructed neural network. The model achieved a global relative error of 0.38 indicating moderate prediction accuracy. Its architecture consists of six neurons in the input layer corresponding to the target elements, three neurons in the output layer referring to the result variables, and four neurons in the hidden layer. In addition, a ‘Bias’ neuron is included at the top of both the input and hidden layers. According to Escandón-Barbosa and Hurtado-Ayala (2014), during the model training phase, this neuron contributes to achieving a more accurate adjustment of the weights for the other variables. Synaptic weights obtained for each connection between layers are shown in Figure 2, and consist of numerical values assigned by the system to connections between neurons. Their function is to regulate the intensity of signals going from input to output. A positive weight value indicates that the neuron gives greater relevance to the information received through that connection. In contrast, negative synaptic weights indicate that less importance has been given to that connection (Escandón-Barbosa & Hurtado-Ayala, 2014) grey, while negative synaptic weights or those less than zero are represented in blue. Based on the weights obtained, it is possible to determine the level of importance of the variables, which is reflected in Figure 3.

Figure 3 depicts the relative importance of each destination attribute as a predictor of the outcome variables. The results show that ‘the quality of food services’ emerged as the most influential attribute in explaining favourable outcomes related to tourist satisfaction and recommendation intentions. This is consistent with previous findings in the literature, which have demonstrated that the quality of restaurant services—including service quality, food quality and the physical environment—, plays an important role in tourist satisfaction and future behavioural intentions (Bichler et al., 2021; Ha & Jang, 2010; Muskat et al., 2019; Nield et al., 2000; Shahzadi et al., 2018). This may be because today’s tourists are not only interested in visiting cultural, historical or natural sites, but are also looking to explore the destination through its gastronomic heritage (Davras & Özperçin, 2021; Dimitrovski, 2016). Therefore, destinations must ensure the provision of an adequate level of quality in food and beverage services, which can represent an opportunity to build a powerful competitive advantage for both the destination and food service companies.

Second and third in importance are ‘the adequacy of tourism resources’ and ‘the tourism resources of the area’. This coincides with the results of the previous descriptive analysis, in which the assessment of these elements achieved the highest average scores. Therefore, the tourism resources available at a destination and their adequacy for facilitating the tourist experience determine satisfaction and recommendation intentions. These findings align with the study conducted by Chen et al. (2016), who confirmed that a positive impression of the destination’s resources positively influences on increasing visitor satisfaction. This is because direct contact with the destination’s resources fulfils certain types of needs or interests (Navarro, 2015). Likewise, these elements are positioned as the main motivating attributes that attract people to a destination (Ritchie et al., 2000). Therefore, tourists who find what motivates them to visit the destination and have a positive experience will result in favourable levels of satisfaction and future behavioural intentions.

4.4.2. Neural Network Analysis for Hotel Category

Figure 4 illustrates the architecture of the constructed neural network and the resulting synaptic weights. The model achieved a global relative error of 0.36 indicating moderate prediction accuracy. The model comprises six neurons in the input layer (corresponding to the target variables), three neurons in the output layer representing the outcome variables, and four neurons in the hidden layer. The figure also displays the synaptic weights, where positive (greater than zero) weights are shown in grey and negative (less than zero) weights in blue. Based on these weights, the relative importance of each variable was calculated, as presented in Figure 5.

Figure 5 illustrates the relative importance of destination attributes as predictors of the outcome variables for the hotel category. The results show that ‘quality of food services’ emerged as the most influential attribute in explaining favourable outcomes related to tourist satisfaction and recommendation intentions. The next most important predictors are ‘tourism resources in the area’ and ‘the adequacy of tourist resources’, respectively. These same elements achieved the highest average scores in the previous descriptive analysis, which reaffirms the importance of proper management of tourism resources in order to achieve favourable results in terms of satisfaction and recommendation intentions.

4.4.3. Neural Network Analysis for Non-Hotel Category

The architecture of the obtained neural network model is shown in Figure 6. The model achieved a global relative error of 0.39 indicating moderate prediction accuracy. The network comprises six neurons in the input layer (corresponding to the destination attributes), three neurons in the output layer corresponding to the outcome variables, and eight neurons in the hidden layer. The figure also displays the synaptic weights, with positive weights (greater than zero) weights shown in grey and negative (less than zero) represented in blue. These weights were used to determine the relative importance of the input variables, as shown in Figure 6.

Figure 7 illustrates the importance of destination attributes as predictors of the outcome variables for the non-hotel category. The results indicate that ‘the adequacy of tourism resources’ emerges as the most influential in explaining favourable outcomes related to tourist satisfaction and recommendation intentions. This finding reveals a difference with the hotel category, in which the most important predictive aspect is ‘the quality of food services’. Therefore, hypothesis 3 is supported in the context of this study. The difference observed between the two categories may be due to people staying in different categories of accommodation establishments may want to do various things, visit different attractions and participate in varied activities while at their destination (McKercher et al., 2023). Hence, the heterogeneity in the needs, interests and perceptions of tourists staying in non-hotel establishments compared to those using hotel accommodation, as determined in this study. These results are in line with the findings of (McKercher et al., 2023), which indicate that preference for certain types of accommodation establishments has some potential to predict visitor behaviour.

Second and third in importance are ‘the quality of food services’ and ‘the tourism resources of the area,’ respectively. When analysing the degree of importance between the two types of accommodation, it can be seen that the quality of food services maintains a relatively high degree of importance in both categories. However, the relevance of tourism resources in the area decreases significantly in the non-hotel category, falling below 40%. This could be because non-hotel establishments, being located in mostly rural areas, have the potential to attract tourists due to the tourist resources they may possess. However, even if there is a variety of tourist resources, if these are not adapted to facilitate tourist visits, tourists are more likely to prefer visiting another destination. Therefore, as indicated by the results in the previous section, it is more important for visitors that tourist resources are suitable for their enjoyment than the variety and presence of these resources in a destination.

5. Conclusions

The results obtained provide a better understanding of how different components of the destination are valued differently depending on the type of accommodation. Although the scientific literature has addressed the accommodation category as a differential factor, the evaluation of destination elements concerning the type of accommodation has been a little explored field of study. For example, authors such as Veloso and Gómez-Suárez (2023) highlight that there are differences in consumer ratings between different types of accommodation, even though the same basic service is expected from all of them. This study also aims to contribute to the literature on destinations, providing a greater understanding of differences in ratings and which attributes lead to greater satisfaction and recommendation depending on the type of accommodation (hotel and non-hotel). To this end, a descriptive analysis and a neural network analysis were carried out, evaluating 47,568 feedback surveys administered by hotel and non-hotel establishments located in Spain to tourists and containing questions about the elements of the destination. Adopting this perspective means obtaining knowledge directly from tourists who have stayed overnight at the destination and who, therefore, have had greater contact with it during their stay. The main finding is that tourists evaluate the destination attributes significantly differently depending on the type of accommodation. This also influences the tourist’s satisfaction and intention to recommend the place visited.

At a theoretical level, this research contributes to the literature by deepening our understanding of tourist evaluations of destinations based on their post-trip experiences, depending on the type of accommodation selected. Taking into account that the post-travel phase influences visit and recommendation intentions (Santoso et al., 2022). In this sense, the elements of the destination that act as predictors of satisfaction and recommendation differ depending on the type of accommodation (McKercher et al., 2023). In the hotel category, ‘the quality of food services’ is the most relevant in influencing marketing results (Pestana et al., 2020), while in non-hotel category, ‘the adequacy of tourist resources’ is the most relevant (Gonzalez-Diaz et al., 2015; Lalicic et al., 2021). These results collectively support Hypotheses 1, 2, and 3, confirming that both the evaluation of destination elements and the impact of those elements on satisfaction and behavioural intentions vary according to the accommodation category.

From a practical perspective, suggestions are provided to destination and tourist accommodation managers, as different aspects of marketing and tourism promotion should be emphasised depending on the type of accommodation. This is in line with the study by Croes (2008), which argues that different types of accommodation target different tourist markets and that it is therefore necessary to diversify marketing strategies. This would increase satisfaction and behavioural intentions for each type, as guests value each aspect differently. For example, in the hotel sector, it is important to reinforce the promotion of food services and emphasise quality, either through the hotel’s own culinary offerings or through partnerships with restaurants in the destination. The emphasis on quality can be operationalized through several mechanisms. For example, by emphasising the provenance of ingredients, collaborating with local producers and suppliers to enhance the sense of place, or participating in gastronomic routes and agri-food initiatives related to products such as cheese or wine, which are traditionally associated with the region and form part of its local identity. Hotels may also organise thematic events or culinary days centred on local products, turning them into an additional attraction that enriches the overall guest experience. On the other hand, non-hotel accommodation should focus on the adequacy of tourism resources which implies highlighting and facilitating access to complementary elements of the destination that enhance the stay. Managers of these establishments can assist guests by providing destination information prior to arrival, including suggestions on activities and attractions to visit. Once on-site, they can facilitate bookings directly, creating synergies with local tourism offices and activity providers. For example, enabling guests to easily book and access local experiences through the accommodation itself can promote stronger collaboration across tourism stakeholders and improve visitor satisfaction.

This research also has some limitations. A specific year was selected for the analysis, so it may not reflect seasonal variations in behaviour. Likewise, in terms of geographical context, Spain was selected as the destination, which may not reflect similarities with behaviour in other countries. In addition, a provider bias may exist, as the data were collected through a single enterprise, which might not fully capture the diversity of users or destinations. Moreover, non-response bias could have influenced the results, since individuals who chose to participate may systematically differ from those who did not, potentially affecting the generalizability of the findings. The use of single-item measures can also limit the results compared to multi-item scales. Furthermore, as all constructs were assessed through self-reported data collected at a single point in time, the study may be subject to common-method bias, which should be considered when interpreting the observed relationships among variables.

Future research could include comparative analyses to assess whether there are differences in evaluations by season and replicating the study in other destinations to assess perceptions in other destinations. Additionally, the analysis could be extended by applying quantitative techniques, such as Partial Least Squares Structural Equation Modelling (PLS-SEM) to gain a deeper understanding of tourist behaviour and to provide further insights into the relationships among variables. In addition, the application of predictive models based on machine learning techniques could offer insights into perceptions and tourist behaviour. Moreover, incorporating data from multiple providers could help reduce potential sampling bias. Mixed-method designs could help capture dynamic changes over time and mitigate common-method concerns. Additionally, employing multi-item and validated measurement scales would enhance construct validity. Although the present study benefits from a large dataset (n = 47,568), ensuring broad representativeness across different destinations and accommodation types, future research could include a more detailed analysis of the regional and typological distribution of establishments to enhance contextual understanding.