1 INTRODUCTION

Tourism, considered as an economic and social activity that promotes and involves the contact tourist-community, can benefit the places wherein it develops, e.g. generation of employment and income, improvements in basic, tourism, and support infrastructures, among others.

According to the United Nations World Tourism Organization (OMT, 2012), tourism accounts for 6% to 8% of global employment and, thus, it can be considered one of the most important industries; for some countries, tourism is fundamental to boost economic growth and social well-being of local populations.

However, if not carefully planned and/ or developed tourism may have negative impacts in the destination and on its community. In the specific case of the religious tourism, as ^{Fagundes (2010}, p. 897) points out, it becomes necessary to consider the activity within a structured and systematic management of the tourism space, both to preserve its characteristics and to mitigate the social, cultural, and environmental impacts, aiming the conservation of the historical heritage of those so-called sacred places; the respect for the local culture; and the balance and equity of their economic growth.

In this perspective, some studies such as those by Eshliki and Kaboudi (2012), Dyer et al. (2007), ^{Gursoy et al. (2002}), Gursoy and Rutherford (2004), ^{Nunkoo and Ramkissoon (2012}), among others, posit that the residents' perception of costs and benefits of tourism influences their level of support for tourism activities in a given locality. It is the tourism managers' responsibility to address that factor aiming to achieve the community support for tourism development.

The general objective of this paper is to analyse the factors which influence the residents' support for religious tourism development in Santa Cruz, State of Rio Grande do Norte (RN), Brazil. Currently, religious tourism in the city of Santa Cruz is increasing boosted by the visits to the 'Santuário de Santa Rita de Cássia' (Shrine of Saint Rita of Cascia), known as the Patron Saint of the Impossible. Santa Cruz is a city located in the State of Rio Grande do Norte, in the microregion of Borborema Potiguar, in the mesoregion of Agreste Potiguar, 115 kilometers from the state capital Natal, RN, with a population of 35,797 according to the 2010 IBGE census (^{IBGE, 2013}). Within this context, it is necessary to properly develop and manage tourism and stakeholders.

2 TOURISM DESTINATION AND STAKEHOLDERS MANAGEMENT

According to the World Tourism Organization UNTWO (OMT, 2001: 38), "tourism comprises the activities of persons travelling to and staying in places outside their usual environment for not more than one consecutive year for leisure, business and other purposes".

It is possible to understand that tourism involves the movement of people from one place to another, for various reasons and for the practice of several types of tourism. It implies, on the part of the tourist/traveler, the availability of time and income, among other aspects. In this sense, tourism planning consists of orienting actions and activities towards a future result. Such planning encompasses either the national, regional, or local space (^{DIAS, 2008}).

Planning is crucial to make tourism more beneficial than harmful to the resident community and other stakeholders. Regarding the concept of tourism destination, ^{Valls (2006}: 15) refers that it can be associated to any territorial unit which has the planning vocation and the administrative capacity to develop it. That means that the tourism destination is a geographical space wherein there is potential and conditions for the management and development of tourism activity.

Within this perspective, the destination management seeks to solve or to avoid problems in a tourism destination. To achieve this goal it should address the economic, cultural, social, political, and environmental dimensions, considering the information and specificities of each one, in order to find a solution for the problems and make the destination sustainable (^{VIGNATI, 2008}).

As to the definition of the term stakeholder, ^{Freeman (1984}: 46) suggests "any identifiable group or individual who can affect or is affected by the achievement of an organization's objectives". In this sense, stakeholder can be an employee of a company, a consumer, supplier, government, or a community that, in any way, can influence, or is influenced by the process of development of a company, organizations, or others. Regarding the classification of stakeholders, these can be classified, in general, as primary and secondary. The primary are the individuals who can affect directly the achievement of a company, such as investors, suppliers, clients, and resident community, among others. The secondary stakeholders are the groups which do not have direct contact with the organization but, depending on the moment, can influence it (^{TEIXEIRA; DOMENICO, 2008}).

Considering the resident community as a primary stakeholder, which can influence the way tourism should or should not develop in a destination, it becomes necessary to include it in the tourism process and consider its perception regarding the impacts (positive or negative) and the benefits and costs of tourism, so that this becomes successful. This same perspective should be considered regarding Religious Tourism.

3 RELIGIOUS TOURISM: Concepts and definitions

Tourism is a comprehensive activity and it can be segmented into more specific types such as Agritourism, Adventure Tourism, Ecotourism, Rural Tourism, Mega-events Tourism, Business Tourism, and Religious Tourism, among others (^{BENI, 1998}). Each niche has specific activities, uses adequate resources, and develops in places which have the conditions or the potential for it

As for religious tourism, according to the Ministry of Tourism of Brazil (MTUR, 2010: 19), "Religious Tourism is the set of tourism activities arising from the spiritual quest and religious practice in places and in events related to institutional religions, regardless of creed or ethnic origin".

In general, religious tourism is the displacement of people motivated by faith to places or destinations considered "holy" or with a strong presence of religious manifestations. In this sense, ^{Beni (1998}: 474) writes that religious tourism,

Refers to a great movement of pilgrims, therefore potential tourists, heading for religious centers, motivated by faith in distinct creeds. This kind of demand has unique features which have led some authors to disregard it in Tourism studies. According to the author, these pilgrims assume a tourism consumption behavior, insofar as they use the facilities and services with a spending structure similar to those of real tourists.

As seen, pilgrims may become potential tourists, insofar as traveling in quest of spirituality does not exclude the use of tourism facilities and infrastructures as hotels, inns, restaurants, among others, similarly to what happens with other segments of tourism. In this perspective, according to ^{Oliveira (2004}: 52),

Religious tourism is not that of religious people or religion. It is tourism motivated by religiosity and religious culture. Therefore, wherever that culture manifests - in the rural, natural, or urban space and whether in everyday life or in festive occasions - the religious tourism (organized or not) may arise.

Despite its motivation, religious tourism still implies the movement of people from one place to another and thus, the use of goods and services in the destination besides those specifically religious.

^{Santos (2011}) argues that the practice of religious tourism entails the realization of activities such as pilgrimages, processions, visit to places considered "holy" (shrines, temples, churches, etc.) participation in festivals of patron saints, spiritual retreats, seminars and other events that emphasize the religiosity and culture in the tourism destinations. Thus, people travelling to destinations where there is religious tourism seek spiritual contact with the divine, whether to thank for a grace, pay a promise, offer ex-votos, or just to visit and get to know the local culture and religion. Still, in this sense, according to ^{Arnt (2006}:21),

The tourism religious format can occur whether it is an individual or an organized travel group engaging in activities such as religious journeys, pilgrimages, or penances, according to the traveler's faith, moral, dogmatic and religious objects. The religious journey happens when the individual, for his own initiative and without expecting material or spiritual rewards, goes to sacred places. The pilgrimage is a visit to sacred places to fulfill a promise or a prior request made to deities or blessed spirits. It is considered an act of penance when the faithful go to a sacred place searching for redemption from sin and guilt by means of religious advice.

In general, religious tourism can be linked to pilgrimages insofar as these imply movement of people motivated by the religious practice or knowledge related to culture and religion. The religious tourist remains a pilgrim, even if they add other services, equipment and infrastructures to their religious travel (^{OLIVEIRA, 2004}). That is to say that the religious tourist, driven by faith, may also have religious-free leisure time and entertainment.

The activities carried out by the religious tourist include routes of a religious nature, visit to shrines in the most diversified parts of the world. ^{Oliveira (2004}: 49) defines shrine as "privileged sites to search for the sacred as a spiritual, mystical, and supernatural dimension of existence. Therefore, the shrines are not necessarily sacred, but a privileged location to experience this sacredness". In short, shrines attract pilgrims/ religious tourists whether for a spiritual encounter with the divine or to get to know the cultural heritage (tangible and/or intangible) of the chosen destination.

Many concerns have arisen with the rapid growth of tourism, in particularly regarding the carrying capacity of shrines and the changes to the landscape caused by the number of construction works. Given this situation the authorities have begun to implement environmental monitoring so as to protect the natural resources (^{FAGUNDES, 2010}).

4 FACTORS WHICH AFFECT THE RESIDENT'S SUPPORT FOR TOURISM DEVELOPMENT

Several studies examine the attitude, participation, support, and others related to the community's perception of the impacts (positive or negative), benefits, and costs which can be assigned to tourism development. Among these studies, ^{Vargas-Sánchez et al. (2011)} proposed a universal model to explain the attitudes of residents regarding tourism development in the Spanish province of Huelva. In that research the following variables were considered: tourism behaviour, density of tourists, and residents' perception of the level of tourism development. These variables are seldom used in similar models.

The findings showed that the perception of positive impacts outweighing the negative is a powerful indicator of the residents' positive attitude towards tourism development. Also, there is a positive relationship between the residents' perception of a respectful tourists' behavior and the attitude regarding tourism development. On its part, the high density of tourists influence negatively the residents' attitude. Likewise, the perceived level of tourism development has a negative influence in residents' attitude, inasmuch as they consider beneficial a moderate development, so, when the development increases, the perception may become negative.

^{Duarte (2007}) emphasized the factors which affect the participation of the community in the development of tourism in Maria da Fé, in the State of Minas Gerais. The author's findings suggest that the local identity, image of tourism, the role of the state, and the view of the contribution of SEBRAE/Minas affected the support and the participation of residents for tourism development. The participation and support, in this case, depended on the residents' perception on tourism activities, the entities responsible for them, and local initiative, i.e. public or private institutions promoting and/or encouraging the community's participation in the tourism process.

Regarding the community support, ^{Gursoy et al. (2002}) put forward a model of causal relationship to examine the community support for tourism development. The study was developed in five municipalities next to a leisure area in the State of Virginia, United States. The results showed that the residents' perception of factors such as their level of interest, 'ecocentric' attitude, use of local resources, and costs and benefits of tourism development can affect the community support for development of the activity.

Subsequently, ^{Gursoy and Rutherford (2004}) added other factors to de Gursoy et al.'s (2002) work, based on the perceived costs and benefits of tourism development. This study was based on the Social Exchange Theory and aimed to analyse the perceived impacts and the factors which can influence people's perceptions and it was developed in fourteen communities from two states of the United States of America. The findings showed that the community support for tourism development is affected directly and/or indirectly by nine factors: residents' level of interest, ecological attitudes, use of community resources, affection for the community, state of local economy, economic benefits, social benefits, social costs, and cultural benefits (^{GURSOY; RUTHERFORD, 2004}).

Therefore, the communities which tend to support tourism development in a given destination are those that perceive more benefits than costs of tourism regarding the environment, social life, local culture, local economy, and political power of decision of each citizen over their customary surroundings.

^{Nicholas et al. (2009}) analysed the factors which influence the community support for sustainable development of tourism in the Pitons Management Area, World Heritage Site, located near the town of Soufriere in the southeast of Saint Lucia, in the Caribbean. There are two volcanos in that town and they are the most visited attractions of Saint Lucia Island. The study showed that the community affection influences the support for sustainable tourism development and the environmental attitudes of the resident community. Also, the study cautions against the lack of involvement of the community in the tourism process and the consequences regarding the sustainability of tourism (^{NICHOLAS et al., 2009}).

To some degree, when a community perceives positive changes and enhancements to its territory due to tourism activities, such as conservation of natural resources, projects and programs of environmental awareness aimed at tourists and residents, among others, it tends to support the development of sustainable tourism; as well, the community tends to change the attitude towards natural resources, since those are finite and both residents and tourists need them to carry out their activities.

^{Nunkoo et al. (2012}) put forward a study on the public trust in the tourism authorities in Mauritius. The results showed that the economic and political performance of the tourism authorities, the power of residents in tourism, and the interpersonal trust determine the trust in tourism authorities, wherein that trust is significantly related to the residents' political support of tourism.

Within the Social Exchange Theory, ^{Nunkoo and Ramkissoon (2012}) tested a model of relationship between variables including the study of the residents' trust, unlike other studies which overlooked the effect of that variable. Such model was tested with the population of the island of Mauritius, located in the eastern Indian Ocean. The authors pointed out that the island's economy is based on traditional activities: farming and textile productions. However, tourism has been developing and is increasingly important for the local economy.

The resultant model proposes that residents' support for tourism development is influenced by the perception of its costs and benefits, as verified in early studies (Nunkoo and Ramkissoon, 2011; ^{Gursoy et al. 2002}; Gursoy and Rutherford 2004, among others), and by residents' trust in government actors (^{Nunkoo et al. 2012}, among others). The model also considers the residents' power of influence on tourism development. The authors conclude by emphasizing the need for further research encompassing the relationship between trust and residents' power of influence on tourism, so that the relationship is confirmed.

In a subsequent work, ^{Nunkoo and Smith (2013}) developed a model of analysis of residents' trust in government actors and political support for tourism. That study was conducted among residents in Niagara, Ontario, Canada. The residents' perception of benefits and costs of tourism and their trust in government actors determine their support for the development of the activity. It is worth noting that residents' perception of political and economic performance of government actors influenced their trust.

In this way, this study aims to verify how the model of causal relationships proposed by ^{Nunkoo and Ramkissoon (2012}) applies to religious tourism in the city of Santa Cruz, RN. It is important to identify the factors of community support so as to achieve a better planning and management of tourism and its stakeholders; promoting stakeholders' inclusion in the tourism process, particularly the residents, who receive tourists, share the infrastructures and, sometimes, suffer the major negative impacts. The methodology used in this study, aiming to achieve the proposed objectives, is presented next.

5 METHODOLOGY

This study is descriptive in nature, followed by a quantitative approach. The descriptive research, according to ^{Martins Jr. (2008}:83) "aims to uncover and observe the existent phenomena, current situations and events, trying to describe, classify, compare, and evaluate them, with the objective of clarifying situations in order to devise future plans and decisions".

The questionnaire was used as a data collection instrument, which for ^{Vergara (2009}:39)

Is a method for collecting field data, interacting with the field composed of a ordered collection of questions regarding variables and situations which the researcher aims to examine. Such questions are presented to respondents, in writing, so that they can also answer in that way, regardless of being presented on a paper or on a computer.

Research took place in the city of Santa Cruz, State of Rio Grande do Norte, Brazil, located 115 kilometers from Natal, the state capital, with an estimated population of 35,797 inhabitants. The territorial unit corresponds to an area of 624.36 Km², with a population density of 57.3 people per Km². In 2010 presented a Municipal Human Development Index (MHDI) of 0.635 (^{IBGE, 2013}).

The city has been trying to develop religious tourism boosted by the visits to the Shrine of Saint Rita of Cascia, which is the main attraction. The Shrine includes a 56-metres high statue of Saint Cascia, chapel, room of the Promises, pilgrims square, auditorium, restaurant, snack-bar, shops, restrooms, viewpoint, and parking lots.

Considering the inhabitants of Santa Cruz, RN as the target population for the study it was necessary a sample of 380 individuals. The data collection instrument was divided in five constructs:

The first construct (factor) refers to the respondent's perception of the benefits generated by tourism development in the region. The second is about the perception of costs generated by tourism development. The third refers to the respondent's perception of residents and businesses' power to influence tourism development. The fourth, refers to resident's trust in government institutions regarding tourism development; and the fifth construct is the respondent's political support for tourism development. The questionnaire was developed based on a five-point Likert scale (Interval from 1 - No benefits to 5 - Many benefits). The scale is not comparative which allows for the respondents to express their opinion more freely in terms of assessing a product or a service (^{MALHOTRA, 2006}).

The data collection instrument was applied to the resident community of Santa Cruz, RN, more specifically to 422 individuals (a number superior to the sample established for this study), from April to July, 2013. The individuals above eighteen years old and living in the city were the target sample.

The theoretical framework of relationship used in this study draws on the model proposed by ^{Nunkoo and Ramkissoon (2012}) composed by manifest variables and latent variables. According to ^{Marôco (2010}:9)

The manifest variables, or observable variables, are variables measured, manipulated or observed directly [...]. The latent variables, Factors or Constructs, are variables not directly observable or measurable, whose existence is indicated by their manifestation on indicator or manifest variables [...].

In this sense, the latent variables of this study are the constructs Perceived Benefits of Tourism, Perceived Costs of Tourism; Power of Influence, Trust in Government Actors, and Political Support for Tourism. The manifest variables are those used in the questionnaire. The description of variables and constructs used in this study are depicted in Table 1. The variables were adapted from ^{Nunkoo and Ramkissoon (2012}).

The statistical package Analysis of Moment Structures - AMOS (v. 18) was used to perform Structural Equation Modelling (SEM) which according to ^{Marôco (2010}:17) "establishes the relationships between variables, latent and manifest, under study".

According to the author, the structural equation analysis can be described as a combination of classic Factor Analysis techniques - which sets a model of measurement which operationalizes latent variables or constructs - and Linear Regression - which set, in the structural model, the relationship between the several variables under analysis.

5.1 Analysis of the Structural Model Obtained

The Structural Equation Modeling (SEM) divides into the sub-models of Measurement and Structural, in which the Structural model, according to ^{Hair et al. (2009}:469), "represents the interrelations of variables among dependency relationships". The analysis of the structural model shows the relationships between variables and constructs thus enabling the identification of which relationships are more representative to the model under study.

Following ^{Marôco (2010}), to test the model fit it is necessary the use of goodness-of-fit indices. Such indices are divided in five broad categories. For this study we used the relative, parsimony, population discrepancy and absolute fit indices:

- Relative Indices: They assess the quality of the model, testing it from the poorest fit to the best possible fit. The three indices used were: TLI - Tucker-Lewis Index; CFI - Comparative Fit Index; and the NFI - Normal Fit Index. The fit is perfect for TLI, CFI and NFI when the values are equal or near 1.

- Parsimony Indices: According to ^{Marôco (2010}:46) these indices "are obtained by the correction of indices with a factor of penalty associated to the complexity of the model". The indices used were the PCFI and PGFI Parsimony. The fit is reasonable when the values are between the interval ]0.6;0.89]. Values above 0.8 indicate a good fit.

- Population Discrepancy Indices: "compare the fit of the model obtained with the sample moments (sample means and variance) regarding the fit of the model which would be obtained in population moments (population means and variance)" (^{MARÔCO, 2010}:46). The index used was the RMSEA - Root Mean Square Error of Approximation. This index presents good fit in the interval [0.05 and 0.08] and very good fit when the RMSEA is inferior to 0.05.

- Absolute Indices: they assess the quality of the model per se without comparing it with other models. The indices used were X²/gl and GFI - Global Fit Index. The index X²/gl considers good fit when the values are inferior to 2 and acceptable when inferior to 5. When GFI presents values between [0.9 and 0.95] indicates a good fit. When GFI is superior to 0.95 the fit is very good (^{MARÔCO, 2010}).

In this sense the values of the goodness-of-fit indices of the model were calculated and displayed in Table 2.

Table 2 Goodness-of-fit indices of the model 

Source: Research Data (2013)

The values of goodness-of-fit indices of the proposed model showed in Table 2 indicate a good fit and a very good fit, except for the parsimony fit indices PCFI and PGFI, which presented a reasonable fit of the model.

Figure 1 presents the structural model obtained with the path coefficients, presented in its standardized form:

Source: Research data (2013)

Figure 1 Structural Model obtained in the study 

The values in brackets represent the error probability when a causal relationship between the respective latent variables is considered. The asterisks in brackets indicate that the relationship between the variables is highly significant. The numbers between latent and manifest variables are the load factors of the respective manifest variables in the construct.

5.1.1 Model Validation

The construct validation was done following ^{Tacconi's (2012}) procedure using composite reliability and average variance extracted. According to ^{Hair et al. (2009}) the composite reliability is a measure of internal consistency of the indicators of the latent variables to the extent that they represent the observed construct. The variance extracted reflects the general amount of variance in the indicators explained by the latent variable (construct). According to the authors, higher values of variance extracted happen when the indicator are truly representative of the latent construct. It is an additional measure to the value of reliability of the construct.

The calculation of these measures is presented below:

Reliability of the Construct = ∑Standard Loads2(∑ Standard Loads )2 + ∑ Indicator Measurement Error 1

Variance Extracted = ∑ (Standard Loads 2)∑ (Standard Loads2) + ∑ Indicator Measurement Error2

The parameters of assessment are, according to ^{Tacconi (2012}), the level recommended by ^{Hair et al. (2009}) and ^{Marôco (2010}), who defined the estimate for reliability of the constructs a value equal or superior to 0.70 and for variance extracted the recommended value should be equal or superior to 0.50.

The results of Composite Reliability of the Construct and the Variance Extracted of the model are presented in the Table 3. From the estimates observed in the table, the constructs BENEFITS, COST and TRUST presented good internal consistency, using, as assessment criteria those recommended by ^{Hair et al. (2009}). In these constructs, the estimated values were superior to the reference values, while in the case of the construct COST, the variance extracted, considering the approximation to the nearest unit, reach the value established as criterion.

Table 3 Value estimate of Composite Reliability and Variance Extracted from the Constructs  

Construct	Variables (Linear Relationships)	Non- Standard Estimate	Standard Error	C.R.	Standard Estimate	p-value	Reliability and Variance Extracted
BENEFITS	BENEF1 ← BENEFITS	1.095	0.090	12.220	0.713	***	Reliability: 0.785 Variance Extracted: 0.549
	BENEF4 ← BENEFITS	1.227	0.096	12.794	0.778	***
	BENEF9 ← BENEFITS	1.000	-	-	0.730	-
COST	COST5 ← COST	1.000	-	-	0.544	-	Reliability: 0.719 Variance Extracted: 0.466
	COST6 ← COST	1.033	0.121	8.548	0.737	***
	COST7 ← COST	1.373	0.159	8.619	0.748	***
TRUST	TRUST1 ← TRUST	1.042	0.062	16.803	0.778	***	Reliability: 0.826 Variance Extracted: 0.547
	TRUST 2 ← TRUST	0.721	0.054	13.344	0.637	***
	TRUST 3 ← TRUST	0.836	0.059	14.264	0.643	***
	TRUST 4 ← TRUST	1.000	-	-	0.874	-
SUPPORT	SUPPORT1 ← SUPPORT	1.000	-	-	0.587	-	Reliability: 0.670 Variance Extracted: 0.404
	SUPPORT2 ← SUPPORT	0.904	0.101	8.908	0.648	***
	SUPPORT3 ← SUPPORT	0.951	0.107	8.853	0.670	***

*** - Highly significant

Source: Research data (2013)

Using as reference these parameters of assessment, the construct SUPPORT does not presents good internal consistency, since the composite reliability was 0.670, inferior to the threshold defined as standard 0.70 and, especially the value estimated of variance extracted of the construct, 0.404, also inferior to the assessment criterion established. This fact can be explained by the low value presented of the load factors of the manifest variables which compose the construct. Next the results obtained in this study are discussed.

6 DISCUSSION OF RESULTS

The findings of this study show a value of 0.49 for the coefficient of determination (R²⁾ of the model, which can be considered adequate for Applied Social Sciences. It means that the latent and manifest variables used in the structural model express, satisfactorily, which factors are able to influence the residents' support for religious tourism development in Santa Cruz, RN.

As for the causal relationships between the latent variables, the construct Perceived Benefits of Tourism presented a highly significant relationship with the constructs Trust in Government Actors and Political Support for Tourism. The construct Trust in Government Actors presented a relationship with the perception of tourism costs, which suggests that such perception can, although incipiently, influence residents' trust, for example, in tourism managers.

Also, the construct Perceived Costs of Tourism does not have a significant relationship with the key-Construct of this research, i.e. does not relate to the construct Political Support for Tourism. In other words, the residents' perception of tourism costs does not affect their decision of supporting or not the development of the activity in the municipality.

In this way, it can be said that the support of the sample of residents is influenced by the perception of tourism benefits and by their trust in government actors, which means that the more the resident population perceive the benefits from tourism, and the more they trust in the managers and governments of tourism activity, the more they are willing to support the development of tourism projects in the city. This fact is confirmed by the works of Nunkoo and Ramkissoon (2011), ^{Nunkoo and Ramkissoon (2012)}, Nunkoo and Smith (2013), among others who point out the significant relationship between the perception of benefits and the trust in government actors, and the community's support for tourism development or tourism projects.

7 FINAL CONSIDERATIONS

The main objective of the current study was to analyse the factors which affect the residents' support for religious tourism development in Santa Cruz, RN, Brazil. In order to achieve that goal, quantitative methods were used, as well as the construction of a structural model representing the local residents' support for religious tourism development.

The theoretical framework allowed for the understanding of some studies which put forward the application of models of relationships between variables, such as those from ^{Nunkoo et al. (2012}), Nunkoo and Ramkissoon (2011), Ko and Stewart (2002), Jurowski and ^{Gursoy (2004}) among others, that added to the discussion of the factors and to the better understanding of the model proposed in this study.

The results suggest that the residents' support for religious tourism development in Santa Cruz, RN is influenced significantly by the perception of the benefits of tourism and by the residents' trust in government actors. The same findings were reported by ^{Nunkoo and Ramkissoon's (2012}) works. Furthermore, significant causal relationships were found between the latent variables Perceived Benefits of Tourism and Trust in Government Actors, and, between those and Political Support for Tourism. However, the variable Perceived Costs of Tourism does not relate significantly with the latent endogenous variable of this study, Political Support for Tourism.

In this way, within this study, the structural model which bests represents the reality of Santa Cruz, RN, regarding the development of religious tourism, can be composed by the following latent and manifest variables: Perceived Benefits of Tourism (generation of employment and income, improvement of the quality of life and general assessment of the benefits from tourism), Trust in Government Actors (Municipal Government; State Secretary of Tourism; Federal Government; and Overall assessment of trust) and Political Support for Tourism (Agreement - Tourism is the most important industry for Santa Cruz; Tourism contributes to the right economic development; Tourism has an important role in the local economic development).

From the structural analysis of the model, a R² value of 0.49 was obtained, which means that for the studied population, the model presented a structural relationship acceptable, and, at the same time it presented goodness-of-fit values between good and very good. Only the parsimony PCFI and PGFI indices obtained a reasonable fit.

Through this research it was possible to investigate how the factors proposed by the model of relationship between variables of ^{Nunkoo and Ramkissoon (2012}) could be applied in the context of the religious tourism of the city of Santa Cruz, RN, Brazil, aiming the inclusion of the resident community in the process of local tourism development.