A behaviour model for cultural tourism: loyalty to destination

Abstract

Tourism in the Spanish economy represents 11% of the GDP (INE, 2018). Its sun and beach tourism model has diversified into other products, such as cultural tourism, that has been increasing in recent years. The objective of this research is to develop a structural model that measures the behaviour of cultural tourists, to help better understand the main variables affecting their loyalty to a destination, travelling within Spain. After reviewing the scientific literature, a hypothetical-deductive method has been used that focussed on the importance of considering the customer's experience, proposes a set of working hypotheses, contrasted by means of an analysis of the structural equations model (SEM), estimated taking the data from the Resident Tourism Survey/FAMILITUR of the National Statistics Institute (INE). The model confirms the importance of socio-cultural variables and the experience of tourists in loyalty-building and the zero importance of the spend in this relationship. This paper analyses the new segments of tourist demand, with a view to the results obtained assisting in the design of differentiated marketing strategies to increase the repetition of the cultural tourisms visits to the same destination and, therefore customer loyalty.

Keywords:

• Cultural tourism
• loyalty
• tourist demand
• structural equation modelling (SEM)
• FAMILITUR
• tourist management

JEL Codes:

• C0
• M3
• Z1

1. Introduction

Intense competition today between the various different tourist destinations to attract travellers (UNWTO, 2019) makes it necessary to implement marketing strategies aimed at different segments of demand and requires specific promotion and policies for their development (Tomić et al., 2019).

To find competitive advantages which both attract and maintain the tourist flow, is essential to position a tourist destination ahead of its competitors. Visitor loyalty to the destination is key in both cases. Maximising tourist retention, encouraging changes of destination, or minimising their loss are the objectives to be achieved (Fornell & Wernerfelt, 1987).

Consumer loyalty is fundamental in the new trends of relational marketing as a key predictor of customers’ post-consumer behaviour customer (Chen & Chen, 2010; Chi & Qu, 2008).

Transferred over to tourism, the increase in the supply of tourist destinations, that broadens the options of choice for visitors, implies tourism management based on the development of stable relationships with tourists who visit them in order to achieve their loyalty to the destination.

The attempt to obtain traveller loyalty through tourist destinations involves understanding how the behaviour of tourists affects loyalty to the destination. The tourist influx to destinations is the result of tourist behaviour (Butnaru et al., 2018).

In the current economic context, the success of tourist destinations depends on the loyalty of tourists (Lončarić et al., 2019). Therefore, the study of tourism demand and behaviour is a priority and the importance of tourism in the Spanish economy is key, already amounting to 11% of GDP (INE, 2018). This implies the need for diversification of its sun and beach tourism model towards other types of products, such as cultural tourism, deriving from the country’s rich cultural heritage.

Thus, it is necessary to expand the study of the behaviour of domestic tourism demand, due to its major contribution to the economic growth of the country (Flores Ruiz et al., 2018). In the case of cultural tourism, as a segment, cultural trips made in 2017 by residents in Spain topped 12.5 million tourists, amounting to a total expenditure of €6.74bn (INE, 2018). Furthermore, research into cultural tourism in Spain reveals underutilisation of a significant percentage of localities with remarkable cultural heritage (Huete Alcocer & López Ruiz, 2019).

Consequently, the goal of this study is to improve insight into the loyalty of cultural tourists to the destinations they visit. This work focuses on identifying the variables that can explain the experience of cultural tourism consumption and its influence on loyalty to the destination. That is, to analyse the value related to the tourist experience (Lončarić et al., 2019; Pine & Gilmore, 1998), and loyalty to the object of consumption: the destination.

Starting from the theoretical orientations that explain the variables that comprise the customer experience (Garduño & Cisneros, 2018; Hombur et al., 2017), e tourist demand data were analysed empirically, along with cultural motivation, provided by the FAMILITUR survey conducted periodically by the Spanish National Institute of Statistics (INE). The analysis of this data is essential to obtain more objective knowledge of the behaviour of the cultural tourist by providing detailed information on both, the number of trips and the profile of the traveller and their motivation for households in Spain.

In order to establish a loyalty model of the Spanish cultural tourist, the scope of the loyalty construct was first defined by determining the variables that comprise it. To do this, we began by measuring loyalty from the indicators taken by tourism studies; the intention to repeat the visit and the recommendation of the destination to third parties (Chi & Qu, 2008; Oppermann, 2000; Prayag & Ryan, 2011; Yoon & Uysal, 2005).

In order to establish what factors, influence the visitor's loyalty to a cultural destination, we analysed some of the main antecedents of loyalty considered in the scientific literature and present in the FAMILITUR questionnaires. We also studied the activities at the destination and the spend, along with other factors less studied, such as the socio-economic profile of the tourist or the degree of organisation of the trip and their causal relationships with loyal tourists. The theoretical model was estimated using structural equation models (SEM).

Therefore, along with the knowledge gathered, this study constitutes a necessary step in providing information to the agents involved in tourism management, to help them develop more efficient relational marketing strategies and build a more significant competitive advantage, thus becoming the desired tourist destination through short and long-term tourism strategies. In addition, the data obtained can assist the development of tourism management policies that will generate social and economic profitability at destinations.

The present study was structured on the basis of a review of the scientific literature on the concept of customer loyalty and its application to tourist destinations, as well as the factors that influence this. Subsequently, the hypotheses that made up the theoretical model were formulated to then analyse the results obtained for discussion and conclusions.

2. Conceptual framework

2.1. Customer loyalty concept

Since the 1920s (Copeland, 1924), loyalty has been studied in specialised literature on consumer behaviour. At present, this is one of the key concepts in relational marketing because of the major importance that customer loyalty has acquired in consumer societies.

In the relational marketing approach, customer loyalty is based on a series of principles focussed on the customer-company relationship. Business competitiveness makes it necessary to adapt to the needs of the demand with strategies aimed at creating value for the customer and with a company vision that integrates the customer within the organisation (Alet, 2000).

Thus, companies, especially in the services sector, direct their marketing policies to achieving long-term customer satisfaction as a means to obtain their loyalty, and devise a strategy to achieve tremendous competitive advantage (Bharadwaj et al., 1993).

However, the concept of loyalty, for other authors, entails more than a relationship between the customer and the company and includes a feeling of adherence to the products/services of this brand and even a positive attitude towards a single specific service provider (Gremler & Brown, 1996).

From the above, it is concluded that the definition of the loyalty construct is comprehensive, and in the literature, there is no one single way to analyse it. Marketing studies, on loyalty, have been advanced from three different conceptual approaches according to the variables used to measure loyalty: the behavioural approach, the attitudinal (Jacoby & Chestnut, 1978; Moore et al., 2015) and an integrative vision of both (Backman & Crompton, 1991).

• The behavioural approach measures loyalty as a behaviour through the number of times a product is repurchased from a brand (Oliver, 1999). Loyalty is analysed as a one-dimensional construct based on the real behaviour of the consumer: the only indicator that is taken into account is the final result based on the frequency of the purchase, measured by the purchase history, without analysing the reasons that enable the service to be contracted again (Tranberg & Hansen, 1986). The recommendation of the product or service to third parties is also a variable used to analyse loyalty (Zeithaml et al., 1996).

• The attitudinal approach is related to the degree of affection; the positive attitude towards the product; or the intention to recommend it (Petrick, 2004) once it has been evaluated by the consumer and the results of the trust it generates appear. To study loyalty, the authors considered the customer's psychological commitment that creates emotional ties with that product, brand or organisation as a result of their experience and that can be expressed as a sense of attachment or affection towards the employees of company products or services (Jones & Sasser, 1995). This approach does analyse the factors involved in the repetition of consumer behaviour. To do this, we proposed a theoretical explanatory model of attitudes, composed of cognitive (the available information or prior knowledge of the brand, product or service that shapes their opinions or perceptions), affective (the positive or negative emotions that can produce a satisfactory or unsatisfactory evaluation by the consumer, regarding the brand or product), and conative attributes (the intention or will enabled us to analyse the predisposition of the behaviour to act), which would influence the achievement of customer loyalty (Oliver, 1999).

• The two previous approaches provide a conceptual framework called conative or compound, which integrates the two visions and implies a broader theoretical orientation of loyalty where the psychological commitment is manifested in an intention of affective repurchase. Loyalty is analysed as a two-dimensional construct integrated by the attitudinal and behavioural. The customer has a positive attitude towards the brand or product that is repeatedly reflected in their purchase. In this type of repeated purchases, other options are not considered by loyal customers, defined as ‘customers who re-acquire the same service provider whenever possible, and who continue to recommend that provider or maintain a positive attitude towards them’ (Kandampully & Suhartanto, 2000, p. 347).

The use of one approach or another, when addressing the study of customer loyalty, depends on the objectives of the research, the available data, or the type of market. The behavioural approach requires information that is easier to obtain, through the purchased data, compared to the cognitive, affective, and conative information needed to measure attitudinal loyalty.

2.2. Components of loyalty to destination

Investigations into consumer loyalty to tourist destinations or products s have been run in recent decades, due to the importance of satisfaction in the commercial positioning of tourist destinations. This satisfaction, as a precedent of loyalty, is derived from the services or attributes of the destination, which materialises into the repetition of the visit or recommendation to third parties. The latter is of great importance as it is one of the sources of information that has the most influence on potential tourists (Gartner, 1993).

In the area of leisure and tourism services, the study of tourist loyalty derives from the concept of customer loyalty, by applying the product category of services and products to the tourist destination (Backman & Crompton, 1991). The consequence of this approach, tourism loyalty, has also been conceptualised following an analysis based on behavioural, attitudinal, or composite variables.

The measurement of loyalty is composed of two dimensions (Bigné et al., 2001; Chi & Qu, 2008; Ozdemir et al., 2012; Prayag & Ryan, 2011; Yoon & Uysal, 2005): behavioural loyalty, evaluating the degree of loyalty for the repetition of the visit (McKercher et al., 2012) and attitudinal loyalty, understood as a favourable attitude towards the destination and measured through positive recommendations – by word of mouth communication - to third parties (Barroso & Martín, 2007).

So, the tourist's loyalty to a destination is expressed through a positive attitude towards a place that later becomes specific in behaviour and implies the repetition of the visit, its recommendation, or both.

Thus, tourism loyalty is built up primarily within the attitudinal dimension, through the positive assessment of the perceived quality of the destination, and that results in satisfaction (Coyne, 1989) and behavioural intentions, and within the behavioural dimension, concludes with the transformation of these intentions into concrete actions to revisit and/or recommend a destination. Attitudinal preferences precede faithful behaviours (Macintosh & Lockshin, 1997). Tourist loyalty will be the result of jointly integrating the variables present in both dimensions (Backman & Crompton, 1991; Iwasaki & Havitz, 1998).

2.3. Behavioural dimension of loyalty to destination

In this study, cultural tourist loyalty to destination, is analysed from the behavioural point of view, measured by the repetition of the visit. This is the indicator used in the FAMILITUR survey to analyse the visitor's loyalty to the destination, taking into account the data collected from the respondents on whether or not this is a first time stay. Thus, a single item is used to explore this construct that can provide reliable information to measure loyalty without lengthening the questionnaire excessively (Oppermann, 2000).

The behaviours of the tourist are the reflection of their loyalty to the destination. In the case of repeated visits, studies on tourist behaviour use this variable to analyse the visitor's loyalty to the destination. The use of this indicator, relative to the number of visits made, derives from the evidence that the acquisition of a brand repeatedly increases the possibility of being repurchased on the next occasion. Thus, tourists can carry out repeated consumption of the destination as a product by repeating the visit, increasing the possibility of a future return of the visitor to that same destination.

This is the case of the study conducted by Oppermann (2000), one of the most relevant studies on loyalty to destination, where 80% of respondents who had made a minimum of five visits to Australia, revisited it later. This highlights the close relationship between the previous and present behaviour of the tourist, confirming that the degree of loyalty of tourists to a destination is reflected in their intentions to visit it again.

There is numerous tourist research focussed on behaviour based on the analysis of the different factors that influence the repetition of the visit to a destination (Alegre & Garau, 2010; Yousefi et al., 2012). In this sense, some authors discriminate between different tourist behaviour by segmenting demand between repeat visitors and first-time visitors (Kozak, 2001).

2.4. Variables that influence tourist loyalty

In this section we shall review the conceptual foundations of the different factors that influence the loyalty of the cultural tourist, focussing only on the variables of the FAMILITUR demand study, analysed for the proposed model.

Several factors can influence the behaviour of tourists as regards repetition of the visit; some of these are derived from the different attributes of the consumer and others arise from the tourist's own experience (including those related to the organisation of the trip, activities at the destination, and expense). There have been several studies on cultural destinations that analyse the positive relationships between the quality of the tourist experience and the loyalty to the destination (Chen & Chen, 2010).

The tourist experience is conditioned by the characteristics of the destination: whether it is emerging or is an already mature destination, and this defines its tourism potential. Several studies show that the ability of the destination to provide the visitor with experiences that meet their expectations and needs will generate a greater or lesser level of tourist loyalty to the destination, manifested both in the intention to recommend the destination and repeat the visit (Bigné et al., 2001).

The tendency to repeat the visit is established when the tourist has been satisfied with the attributes of the destination during their first visit (Kozak, 2001). Boo et al. (2009) link the attributes of a product that are related to the perceived value in marketing, transferring them over to the attributes of a destination and its perceived value as a fundamental criterion in loyalty to the destination.

For this reason, the possibilities offered by the destination in terms of activities in general, and the cultural tourist offer available at the destination, are essential in loyalty-building. If the possibilities of involvement int new activities at the destination have been exhausted, it may be that, although the tourist experience has been positive, the visit will not be repeated. However, if tourists e can enjoy new experiences on the next visit, or repeat the same activities they find attractive, they may revisit that destination.

Consequently, the following hypothesis is proposed:

Hypothesis 1 (H1): The activities carried out at the destination have a positive impact on loyalty to it.

Regarding the expenditure incurred before and during the journey, scientific literature shows that tourism expenses are determined by factors such as psychographic aspects of the tourist (personality, lifestyle, interests, hobbies and values), their primary motivation to travel, or travel characteristics.

The influence of spending and the repetition of the visit to the destination have been analysed in some scientific studies that establish a relationship between the number of visits made to the destination and different spending patterns. Thus, the repetition of the visit may involve more or less expense, through the search for different experiences when repeating the stay (Lee et al., 2015).

The socio-cultural profile of the tourist would also affect spending directly. Their level of education, which makes them an "active" tourist, eager to interact with the destination (heritage, local population, etc.) and their purchasing power, which, in the case of the cultural tourist is medium-high (De la & Varquero, 2003); combine to provide a profile of a consumer willing to hire certain quality tourist services.

As a result of the above, it has been decided to try to establish the relationships arising from the different spending options at the destination and loyalty to this, through the following hypothesis:

Hypothesis 2 (H2). Spending before and during the trip influences loyalty to the destination.

The degree of organisation of the trip also affects the experience of tourists at the destination. The data provided on this by the survey used here (FAMILITUR) shows that the cultural tourist makes little use of standard, organised trips (tour packages), and this survey of internal demand for cultural tourism reflects less use of tourism packages and a more significant independent organisation of the trip. It is also explained by the fact that the purchase of these tourist products increases with increasing physical and/or cultural distance between the point of origin and destination (De la & Varquero, 2003). This would have an impact not only on the share-out of tourist spending at the destination but also on the way of actually experiencing the destination visited.

The composition of the tourist group also influences the repetition of the visit. Thus, Campo et al. (2010) have analysed the type of travel groups on the island of Mallorca, showing that families had a stronger intention of returning to the destination than groups of friends who perceived a much more negative image of the destination.

Consequently, the following hypothesis is proposed:

Hypothesis 3 (H3). The organisation of the trip positively influences the loyalty of the cultural tourist to the destination.

Other factors that influence tourist loyalty may derive from the different attributes of the tourist. Thus, the socioeconomic, demographic profile or personality would influence the degree of consumer loyalty (Li et al., 2008). The place of urban or rural residence, age and gender would be some of the variables that modulate tourists’ level of loyalty (Petrick & Backman, 2001).

Correia et al. (2015) link older tourists with a higher probability of repeating the visit, compared to younger ones. Furthermore, tourists’ intention of revisiting the destination decreases as their purchasing power increases.

Oppermann (2000) also establishes a close relationship between socioeconomic, demographic, and psychographic variables and their loyalty to destination.

According to the literature review, the following hypothesis is formulated:

Hypothesis 4 (H4). The socioeconomic profile of the tourist influences loyalty to the destination.

3. Proposed conceptual model

Taking into account all the above, what is intended with this study is the proposal and analysis of a structural model to measure tourist behaviour, where the primary motivation is cultural. This helps us to ascertain and gain a better understanding of their loyalty to the destination.

The model has been estimated and the proposed hypotheses empirically tested with the data provided by the FAMILITUR survey conducted by the INE, using an approximate monthly sample size of 16,400 surveys, for the study of tourist trips and excursions by the population resident in main family dwellings in Spain for the period between February 2015 and September 2016. This database offers detailed information on the number of trips as well as the profile of the traveller and their motivation (Prado-Mascuñano, 2013).

Table 1 shows a summary of their main characteristics.

Table 1. Information about the sample characteristics.

Type of survey	It continues every quarter.
Population reach	Population over 15 years old residing in the main family home.
Ambit	All the national territory.
Reference period	Monthly.
Sample size	Around 16,400 interviews conducted each month.
Information Collection	Telephone interviews and, in some cases, personal interviews.

Source: National Statistics Institute (2018).

Based on the theoretical foundations set forth in scientific literature, the proposed conceptual model was designed to identify the most significant elements and propose the system of interdependencies that relate these.

To do this, we firstly identified, the variables of the Resident Tourism Survey, FAMILITUR, that are considered to influence the behaviour of cultural tourists at destinations, based on our review of the literature

Thus, a total of 54 variables are included in the exploratory model: 9 related to the socio-cultural profile; 23 to the organisation of the trip; 6 to the activities participated in at the destination; 15 to the expenditure involved, and one variable to loyalty.

Next are presented the factors that would determine the behavioural model, whose structure is to be contrasted through the relationships established in the proposed causal model. The resulting factors of the model were the following:

• Factor 1. Socio-economic and cultural profile of the tourist, composed of nine variables or indicators: age, sex, educational level, professional status, economic activity, household income, characteristics of the location of residence, and number of members that made up the household.

• Factor 2. Trip organisation, consisting of twenty-three variables grouped according to the destination, type of trip organised, services used at the destination, and the reservations made.

• Factor 3. Activities participated in at the destination, consisting of six variables: Cultural visits, attendance at cultural performances, other cultural activities, city visits, visits to rural destinations, and gastronomic activities.

• Factor 4. Expenses incurred, consisting of fifteen indicators grouped by expenses incurred before and during the stay and their amount.

• Factor 5. Loyalty to the destination or type of destination, measured by repeating the visit.

The four latent factors or variables that compose the socioeconomic profile, travel organisation, activities participated in at the destination and expense would, according to the theoretical model, have a direct effect on loyalty to the destination and type of destination.

As a result of the above, four hypotheses have been posed regarding the positive impact on the cultural tourist's loyalty of their demographic and social characteristics; the tourist experiences; level of expenditure, and the way of structuring the trip.

4. Methodology

This research has assessed the effect of variables considered as causes (independent or exogenous variables) upon another variable, considered as effect (dependent or endogenous variable). This way, the research framework used aims to demonstrate the causal relationships between the following five latent factors or variables related to the socio-economic and cultural profile of the tourist; the organisation of the trip; the activities participated in at the destination; the expenses incurred; and the loyalty to the destination or type of destination.

The estimation of the model has been based on the application of the PLS method (“Partial Least Squares-Path Modelling”) to analyse the different hypotheses raised in the study and the total adjustment of the theoretical model. Initially conceived by Wold (1980) as an analysis model within the framework of structural equation models (SEM), it is especially useful in situations where researchers intend to analyse relationships between latent variables in complex models, in research oriented to prediction and where knowledge and theory are relatively scarce.

It is thus a very useful tool to respond to one of the purposes of empirical research: the discovery of causal relationships between concepts - even in the area of social sciences, where it is necessary to use indicators to establish relationships.

In tourism, numerous empirical studies have been carried out, using structural equations, to establish the causal relationships between different factors and loyalty (Brandano et al., 2019; Lv et al., 2020).

Besides, these types of models would be more robust due to deviations from some of the usual assumptions in the models of classic structural equations, such as multivariate normality, or the need for bigger sample sizes.

In particular, PLS has an advantage over all other methodologies by not requiring that the distributions be normal or known. To clarify the relationships between the variables of the model and, given that complex hypotheses are required, curvilinear effects are used to test the full range of relationships between the factors using the Warp 3 algorithm (Temme et al., 2006), using the WarpPLS 6.0 software (Kock, 2017). This algorithm attempts to identify relationships between latent variables whose derivatives are U-curves. This way, we can model not only non-linear effects but also, simultaneously, effects of mediation and moderation, which constitute a much closer approximation to reality.

To assess the adequacy of the theoretical model concerning the data collected in the study sample, different parameters were analysed: on the one hand, the overall adjustment of the total theoretical model, through the evaluation of different adjustment indices. That is, to what degree the model is an adequate representation of the pattern of relationships that exists between the data of the set of partial regressions for each effect of the model.

Cut-offs of each index value, currently considered as suitable markers of the model, can be assumed to be correct and are presented in Table 2. These values have been proposed only recently, in parallel with the best known SEM model indices, based on the covariance of all variables, however the data in recent studies support their usefulness (Henseler & Sarstedt, 2013).

Table 2. Overview of the indexes of fit and reliability of the model.

Index	Value	Value Interpretation
Average path coefficient (APC)	APC = 0.161, P < 0.001	Significant if p < 0.05
Average R-squared (ARS)	ARS = 0.208, P < 0.001	Significant if p < 0.05
Average adjusted R-squared (AARS)	AARS = 0.208, P < 0.001	Significant if p < 0.05
Average block VIF (AVIF)	AVIF = 1,278	Acceptable if < = 5, ideally < = 3.3
Average full collinearity VIF (AFVIF)	AFVIF = 1,437	Acceptable if < = 5, ideally < = 3.3
TenenhausGoF (GoF)	GoF = 0.281	Small> = 0.1, medium> = 0.25, large> = 0.36
Sympson's paradox ratio (SPR)	SPR = 1,000	Acceptable if> = 0.7, ideally = 1
R-squared contribution ratio (RSCR)	RSCR = 0.1,000	Acceptable if> = 0.9, ideally = 1
Statistical suppression ratio (SSR)	SSR = 1,000	Acceptable if> = 0.7
Nonlinear bivariate causality direction ratio (NLBCDR)	NLBCDR = 0.600	Acceptable if> = 0.7

Source: WarpPLS 6.0.

5. Results

In Figure 1, the diagram of the proposed model is presented graphically with the values on the arrows that represent the value of the estimated parameters followed, in brackets, by their corresponding p-value and including the value of the coefficient of determination (R2) for the effect represented.

Figure 1. Model results specifying the coefficients of determination.

Source: WarpPLS 6.0.

This figure interprets the values and signs resulting from the estimates of the coefficients of the structural equations that would explain the causal relationships (impacts) of the different latent variables present in the model. This analysis contrasts the theoretical hypotheses of the structural equation model.

The assessment of the structural model of the four hypotheses posed supports three hypotheses, accepted with a confidence level greater than 95%, p < 0.05. Hypothesis 4 (H4) could not confirm to the confidence level of 95%, this being slightly lower, to 71%, so that this hypothesis is supported.

The multiple squared correlations (R²) is .011 for loyalty, which indicates that the profile of the tourist, the organisation of the trip, the activities and the expense explains11% the behaviour of this factor.

The beta coefficients obtained show the following relationships of the hypotheses posed in the research model. Thus, the following indicators that exert an influence, from lower to higher order, loyalty are as follows:

• The expense (beta = 0.00). There is no relationship.

• The tourist profile (beta = 0.07).

• Activities participated in at destination (beta = 0.12).

• The organisation of the trip (beta = 0.26).

The coefficients show that it is the organisation of the trip, which has the greatest power to predict the degree of loyalty of the cultural tourist, followed by the activities carried out at the destination. The relationship of the socio-economic profile of the tourist with loyalty is weak. It should be noted that the level of expenditure has no influence on the loyalty to the destination.

In summary, the results for each hypothesis are presented:

H1. The activities participated in at the destination have a positive impact on the loyalty to the destination (β = 0.12, p <.01). Hypothesis confirmed.

H2. The expenditure before and during the trip influences the loyalty to the destination (β = 0.00, p = 0.29). Hypothesis not supported. The expense made before and during the trip does not significantly affect the loyalty to the destination of the tourist.

H3 The organisation of the trip positively influences the loyalty of the cultural tourist to the destination (β = 0.26, p <.01). Hypothesis confirmed.

H4 The socioeconomic profile of the tourist influences the loyalty to the destination (β = 0.07, p <.01). Hypothesis confirmed.

Let us stop to analyse each one of these.

H1. The activities participated in - “experiences” at the destination - have a positive impact on the loyalty to the destination.

The relationship between these two variables is positive and statistically significant. In this case, the graph shows a linear function: higher scores in the activities participated in are associated with greater loyalty for all activity levels (Figure 2).

Figure 2. Hypothesis 1.

Source: WarpPLS 6.0.

H2. Spending before and during the trip influences the loyalty to the destination.

The relationship is practically null and statistically not significant (p = 0.29). The curve in Figure 3 is an almost inverted “U” shape for the association of these two latent variables. Thus, medium-low levels of expenditure are associated with higher loyalty and medium-high levels are associated with lower loyalty, however, there is no relationship between the two variables for average levels of expenditure.

Figure 3. Hypothesis 2.

Source: WarpPLS 6.0.

H3. The organisation of the trip influences the loyalty of cultural tourists to the destination.

It can be assumed that this hypothesis is supported, with a medium-low regression coefficient size, but positive (0.26; p <.01). The graph shows that the function is almost linear, indicating a systematic increase in loyalty scores for higher organisational scores (Figure 4).

Figure 4. Hypothesis 3.

Source: WarpPLS 6.0.

H 4. The socio-economic profile of the cultural tourist influences loyalty to the destination.

The weight of the association between these two variables is reduced, although positive (0.07). According to the function shown below, this positive effect is more marked and practically a linear increase from medium-low scores in the socioeconomic profile. Therefore, in the range of very low scores in the profile, there is no association with loyalty. Nonetheless, from those scores, each increase in the level of socioeconomic profile implies a continuous increase in the level of loyalty (Figure 5).

Figure 5. Hypothesis 4.

Source: WarpPLS 6.0.

6. Discussion and conclusion

The results analysed in the previous section show the importance of the type of organisation of the trip, expressed by the cultural tourist: the socio-economic characteristics of the customer; and the activities participated in, on behavioural loyalty, measured by the repetition of the visit to that destination. As well as the null impact of spending on loyalty.

The relationship between the organisation of the trip and loyalty to the destination (H1) is a hypothesis supported by a high degree of compliance and confirms findings from previous analyses (Bigné et al., 2001; Boo et al., 2009; Kozak, 2001). Therefore, those responsible for marketing cultural destinations must consider how the cultural tourist organises their trips. The Resident Tourism Surveys/FAMILITUR evidences the little use of the vacation package and the “do-it-yourself” independent organisation of the trip in the majority of cases. Thus, tourism services should be reinforced, along with resources that could offer more celebrated attractions and respond to the needs arising from this way of organising trips, thus building the loyalty of the tourist to the destination.

Regarding repetitiveness, a negative causal relationship with spending (H2) can clearly be seen. Therefore, marketing strategies for destinations must consider that certain types of high-cost tourist products or services do not result in more frequent repetition of the visit. De la and Varquero (2003) and Lee et al. (2015) also refer to this differentiation in tourist-type.

The relationships between the activities carried out (H1), and the sociodemographic profile (H4) with loyalty are supported hypotheses. These results agree with the ones obtained by Oppermann (2000), Petrick and Backman (2001), Li et al. (2008) and Correia, et al. (2015). In the case of the activities participated in, the confirmation of the hypothesis suggests the importance of the diversity of tourist attractions and supplementary offers of destinations that encourage cultural tourists to return to the destination.

6.1. Contribution of the study

In short, planning and management of cultural tourism destinations must include identifying customers’ needs; improving the orientation of the tourism product; and thereby generating strategies for loyalty to cultural destination to assist tourism professionals and all involved in the sector

6.2. Practical implications

The identification of different variables that affect cultural tourist loyalty to a destination could lead to specific policies for promotion and marketing of cultural destinations, adapting marketing strategies to the cultural pattern of this demand segment.

6.3. Limitations of the study

The limitations found in this research have arisen basically from the variables in the FAMILITUR Survey, thus the questions and answers posed certain limitations.

Variables are framed within the general guidelines given by the WTO (World Tourism Organisation) for studying demand, without conducting a more operational adaptation of some variables. This is the case of the loyalty construct that incorporates only one item, when viewed from the behavioural point of view, as a repetition of the visit, and not attitudinal. Consequently, that exploitation of this statistic establishes a general framework that is not as specific as would be required for further study on segmented demand, such as cultural tourism flows.

This implies that, for a more in-depth approach to the analysis of this type of tourism demand, integrated concepts or variables that are more operational, and adapted to the needs of the field studied, would be needed. In this case, cultural tourism, providing specific data on the different segments of demand, could provide an opportunity for differentiated marketing.

6.4. Future research opportunities

To complete an even broader the study of the behaviour of cultural tourists in Spain in the future, it would be advisable to develop the following lines of research by extending the analysis of the relationship between cultural tourist satisfaction and future intentions to repeat the visit, as well as the influence of these relationships on the recommendation of the destination to third parties. In addition, the model proposed l for the cultural tourist could also be applied to other types of tourists and destinations, such as health, nature or sun and beach.

Acknowledgments

Thanks to ESIC Business & Marketing School for the support of this research.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

Data set associated with the paper will be provided on demand.