An Empirical Investigation of Destination Branding: The Case of the City of Rio de Janeiro, Brazil

Abstract

This paper conducts a study on brand image formation. In this sense, this research was developed to contribute to the tourism sector, and to do so, it had two major aims. The first one was to evaluate if destination branding factors influence the image assessment of a specific tourist destination: the Brazilian city of Rio de Janeiro. The second aim was to estimate if the image of this tourist destination is directly related to loyalty. Therefore, this paper provides a theoretical and an empirical contribution to the theme by using a set of destination branding dimensions to evaluate the image of Rio de Janeiro and to test its influence on destination loyalty. This research follows a quantitative exploratory approach, and in order to reach the designated aims, two research techniques were used: exploratory factor analysis and confirmatory factor analysis. The results confirmed four out of the five hypotheses established a priori in the proposed model. It can be concluded that the structural model is a reliable model.

1. Introduction

Tourism is a sector that grows and expands rapidly, and by doing so, it moves economies, communities, and organizations worldwide. In this scenario, there is an ongoing concern on building a destination brand with a strong and representative image. This way, this study was developed from the idea of image formation through destination branding.

A brand is a combination of tangible and intangible attributes. It is a product or a service that is preceded by dimensions and aims at consumer satisfaction, which, by its turn, is materialized through functional, emotional, self-expression, and social benefits [1,2,3]. In this context, it is worth pointing out that destination branding is one of the theories that guide the brand assessment of a tourist destination.

Destination branding is an important marketing strategy tool for a tourist destination since it emphasizes the differences between destinations, and as a consequence, it creates market opportunities. The image projected by a brand becomes an essential element to establish ideal and effective marketing strategies by demonstrating its unique characteristics, which can become an opportunity to face the competitive market [4,5,6,7,8,9].

In the scientific theoretical context, destination branding is a theory of brand measurement that proposes a theoretical model composed of several factors, such as destination attributes, brand equity, brand personality, brand as a symbol, image, and loyalty [10,11,12,13,14]. These factors, identified in the literature, give support to the development of this study.

These factors also serve as support to measure the image of the tourist destination. The image has an impact on the tourist decision-making in choosing a tourist destination. In this way, the image is important for the creation of public policies that contemplate aspects such as culture, security, and the natural attributes of the destination to guarantee the sustainability of the tourist destination [15,16,17].

It is important to highlight that this paper has two major aims. The first one is to evaluate if destination branding factors influence the image assessment of a specific tourist destination: the Brazilian city of Rio de Janeiro. The second aim is to estimate if the image of this tourist destination is directly related to loyalty. Therefore, this paper provides a theoretical and an empirical contribution to the theme by using the destination branding dimension set [13] to evaluate the image of Rio de Janeiro and to test its influence on destination loyalty. This study is crucial for managers since it creates opportunities for tourism marketers’ specific paths to influence tourists’ image evaluations by means of destination branding.

This research has a quantitative exploratory approach, and, in order to reach the designated aims, the following research techniques were used: exploratory factor analysis (EFA), confirmatory factor analysis (CFA), and structural equation modeling (SEM). Therefore, first, the theoretical framework and the hypotheses of research are presented. Second, the materials and methods used are described. The third part presents the empirical results of the proposed model. The fourth part reveals the theoretical and managerial implications, and the conclusions regarding the study.

2. Theoretical Framework and Research Hypothesis

2.1. Destination Attributes

Attributes are the characteristics of a product or a destination. In this study, they are associated with the idea of destination branding. As Aeker, Keller, and Carvalho point out, it should be noted that the tourist’s perception of attributes shows what factors were determinant for a destination choice [13,15,18]. Also, for Pike, from a marketing point of view, attributes represent the offers for the tourist, which include natural, commercial, and non-profit resources, facilities, amenities, and people [19,20].

For Pike, these attributes are not controlled by the destination [19]. On the contrary, they are market offers, but it is up to the managers of the destination to organize the markets, so they can meet the needs of the tourists and also of the local population. As Tasci and Denizci state, it should be pointed out that the target image is a factor that influences the final decision of the client [21]. Also, as shown in the literature, it is important to notice that there still are gaps that must be taken into account so that new studies can be proposed.

The variables presented in the factor of destination attributes were identified in Carvalho’s doctoral thesis and adapted by the author who followed Aaker’s book “Creating and Managing Successful Brands” [13,15]. In order to compose these variables, Pike and Ryan’s studies were also important [22]. In these studies, the authors give greater attention to cognitive, affective, and conative aspects. In this sense, Pike and Ryan bring, in their scale, attributes that are linked to the quality of service offered in a study predominantly quantitative [22]. The authors argue that a tourism destination positioning depends on a succinct, focused, and consistent message, whereby they place attributes as part of the message.

In addition, it should be highlighted that, in this paper, attributes are organized from Pike and Ryan’s functional and psychological components (i.e., landscape, natural resources, costs, prices, climate, tourist places, activity, nightlife, fun, sporting facilities, activities, local habits, culture, and cuisine and gastronomy) [22]. Also, in this study the scales were crossed and resulted in the association of the destination attributes’ relative branding. Given the above, it is possible to create a hypothesis: H1—Destination attributes influence the destination image.

2.2. Brand Equity

A brand is an element that expresses a value to its consumers. In this sense, according to Aaker, a brand equity estimation “begins by determining the value of business units in the market of products impacted by the brand” [3]. Brand equity is represented by the brand performance and it is measured in terms of a financial value in the company balance sheet [23,24].

Brand equity encompasses five constructs that are related to consumer perception: brand loyalty, brand awareness, perceived quality, brand associations, and other company assets. It is worth pointing out that for Aaker, brand equity impacts both the value to the customer and the value of the company [15].

Kapferer argues that brand equity is the ability of brands to deliver profits [25]. The author also points out that a brand has no financial value unless it can generate profits. In order to say that the lack of profit is not a brand problem, but a business problem, it is necessary to separate the brand from the business—an intellectual temptation.

The concept of brand equity boils down to the idea that if a product, a service or a destiny (as in the case of this study) acquires a positive, powerful, and solid reputation, it becomes a valuable resource—sometimes more valuable than all of its tangible assets because it represents the ability of the place or the organization to continue trading with a healthy margin, provided that its brand remains intact [26]. H2—Brand equity has an influence on the destination image formation.

2.3. Brand Personality

Brand personality tends to be a driver for the representation of the brand functional benefits and attributes. Therefore, to consolidate a strong personality it is necessary to provide energy to the brand by adding interest and involvement. Thus, the personality factor will be able to increase the brand experience and perceptions in a positive way [3,27].

Also, it is important to highlight that brand personality has an association with human characteristics and serves as a symbolic and self-expression function. In this sense, it is capable of embodying a product, thus becoming a part of the brand image [28,29]. Thus, in exploring the characteristics associated with the human being it is possible to unveil the emotional response that the brand can provoke [3].

With regards to the questions elaboration that are related to brand personality, we had as a basis the scale of Aaker that was replicated in the study by Carvalho [10,13]. The author’s study aimed at developing a framework of brand personality dimensions and a reliable, valid, and generalizable scale for measuring the dimensions. The results suggest that consumers realize that brands have five distinct personality dimensions: sincerity, excitement, competence, sophistication, and robustness. H3—Brand personality influences the destination image formation.

2.4. Brand as a Symbol

Carvalho believes that the brand as a symbol can be very powerful because it can have the capacity to provide cohesion and structure to a brand identity, benefiting the recognition as well as the brand recall in the consumer’s mind by stimulating the memory [13]. The author also points out that the existence of a symbol is a fundamental component for the brand development. This way, its non-existence can constitute a substantial deficiency since the symbol represents the symbolism of the identity potential power.

In Keller’s view, symbolism encompasses several brand elements that characterize brand identity. The author also highlights that these identity symbol-forming elements are the name, the logo, the symbol, the slogans, the jingles, and the packagings [28]. H4—Symbolism influences the destination image formation.

2.5. Image

With regard to the destination image factor, it is essential to be aware that the image is the sum of the global product offered plus the set of communicative actions that this product or destination is subject to. Thus, the image must be analyzed by the way the consumer perceives it [30].

The continued development of the customer’s relationship with the brand is the base of the brand. In this relationship, the customer forms a differentiating image of a physical asset or service [13]. As far as the brand image is concerned, Grönroos states that its role is to make filters in the communication and recommendation effects [31].

During all the stages of destination selection for a trip, tourist images help to determine which destinations remain and which ones will be eliminated from the final choice [32]. Therefore, understanding how tourism images are formed can assist destination managers and organizations in developing destination images suitable for selected target markets [33].

Differentiating the offer is a fundamental part of destination marketing. In such manner, working the image properly and efficiently has become critical in strategic marketing. Studies have been arguing that destination image has become a key element in destination marketing aiming at a differentiation feature in order to make the destination competitive. Thus, it becomes increasingly necessary to understand the destination image formation, not only for professionals, but also for scholars [34,35].

Carvalho states that a striking feature of tourist destinations is that they compete among themselves [13]. Buhalis points out that, in this competitive scenario, image has become a decisive factor in the purchasing decision process of the “tourist destination” [36]. Pike, by his turn, points out that the image formation process needs to be worked efficiently, within the marketing area, since it is intrinsically linked to consumer buying behavior and decisions [19].

Corroborating this perception, Moraga, Artigas, and Irigoyen point out that the destination image plays a fundamental role in the process of tourist choice [6]. Over the years, it has become a tool to establish marketing strategies that consolidate the destination in a way that makes it competitive. Image is also a strong influencer of customer attitudes, which can intervene in the destination loyalty [37].

With regard to the satisfaction related to the destination image, Chi and Qu point out that the destination image plays a positive role influence on the quality and satisfaction perceived [38]. In this sense, the authors highlight that the image derives from positive experiences. In other words, tourism satisfaction improves the destination image. The authors also state that the destination image affects the tourists’ behavioral intent. Bearing in mind that a favorable image will influence the destination loyalty, it increases the probability of the tourist return.

On one hand, some authors that make short notes about the image factor, such as Crompton, state that the scope of destination image can be greater when worked in conjunction with the brand [39]. Bigné, Font, and Andreu, on the other hand, consider that the destination image is an important factor in generating satisfaction among tourists and residents [40].

In his study, Carvalho identified that there are consolidated relationships between satisfaction and destination quality through the image factor. In this respect, the author corroborates the view of Pike and Ryan, who, through their theoretical studies and surveys, concluded that the destination image has a direct influence not only on behavior but also on consumer or tourist satisfaction [13,22].

Through a literature review exercise, Chiu, Zeng, and Cheng concluded that the destination image is one of the most important travel antecedents for the pre-tourist because it influences the final purchase decision, and also, the tourist behavior during the trip [41]. With regard to image formation, the author points out that it is based on the interaction of the cognitive and affective components. In his study, Chiu, Zeng, and Cheng confirmed the relationships that were already consolidated in the literature, in which affective and cognitive components are directly related to customer perception and satisfaction [41]. H5—Image has an influence on loyalty.

2.6. Brand Loyalty

Loyalty is an element that becomes critical when value is attached to a particular brand, which, by its turn, is sold and bought. In this sense, it is expected that the brand generates a base of loyal consumers, thus configuring its loyalty, generating less vulnerability and more consistency from loyal consumers [10,42].

It is noteworthy that the similarity of the concept of loyalty is recurrent in tourism studies. Thus, loyalty is measured by analyzing the intention to continue buying the same product, the intention to buy more of the same product, to repeat the purchase, and the willingness to recommend the product to other [3,13,19,38,43,44,45,46,47,48,49].

According to Lübeck et al., “the meaning of the construct Loyalty goes beyond the behavioral dimension (repurchase) and it is necessary to take into account psychological meanings.” [50]. So, loyalty does not only mean customer loyalty, but also that companies or organizations must be faithful to their customers, because when both parties win, a relationship is formed [13]. Carvalho emphasizes that “when a given brand manages to make the customer feel connected to it and when the offer of that brand is different from the competing Services, it can be expected that a favorable recommendation effect will follow”, that is, when the client is motivated, he/she will speak positively on behalf of the brand [13].

Carvalho points out that attitudinal loyalty is linked to satisfaction and quality [13]. The author, then, works with attitudinal loyalty in his study. Uncles, Dowling, and Hammond highlight that loyalty is primarily an attitude that tends to lead to a prolonged relationship with the brand. In this paper, loyalty is the output construct of the structural model [51].

According to this theoretical framework, five hypotheses were proposed regarding the city of Rio de Janeiro, Brazil. The proposed theoretical model is presented in Figure 1.

3. Materials and Methods

This research was carried out with a quantitative approach using statistical methods, which are based on the statistical theory of probability which constitutes an aid to research in the area of social sciences [52].

The research tool used to collect the empirical data included demographic data and the scales for measurement of the destination attributes (with 19 items), the brand equity (with 4 items), the brand personality (with 5 items), brand as a symbol and image (with 6 items), and loyalty (with 4 items), adapted from the studies of Carvalho, Pike and Ryan, and Aaker [10,13,22]. All items were measured using a 7-point Likert scale, ranging from 1 “totally disagree” and 7 “fully agree”.

The sample chosen for this study was non-probabilistic, which, according to Mattar, "is the one in which the selection of the population elements to compose the sample depends, at least in part, on the researcher judgment or on the interviewer in the field" [53].

The data collection was made in a social network, using the snowball collection method, which consists of sharing the research with subjects who have a set of knowledge and friendship relations between them, called a “social network”. The search result had a total of 767 respondents. The sample size was considered by the authors as representative. This is because the research sample was larger than the sample of 385 respondents, if the research was probabilistic (calculation considering a sample error of 5%, confidence level of 95%, and a population of 9.3 million tourists) [54].

The methods used were descriptive statistics of the percentages of respondents that delineated the socio-demographic profile. With regard to the statistical methods used, they consisted of EFA, CFA, and SEM, which were based on theoretical analysis of Hair Jr et al.

The hypothesis model proposed in Figure 1 was adapted from Carvalho, aimed at meeting the main goal of this study, which was to estimate if the factors present in the composition of the destination branding model were able to evaluate effectively the image and the loyalty to the destination [13].

4. Results

4.1. Descriptive Analysis

The study sample consisted of 678 valid responses—64.1% of the respondents were female, 35.5% were male, and 0.4% were non-binary. With respect to the respondents ages, the majority were between 26 and 35 years old (45.2%), followed by those between 18 and 24 years old (25.3%). Also, it was observed that a considerable portion claimed to have postgraduate education: 39.4% said they have completed their Master’s degree and 14.2% have PhD. It is also worth mentioning the respondents’ place of origin: 40.7% are from the southeast region, 34.4% are from the south, 9.2% are from the northeast, 6.5% from the central-west, 2.9% are from the north, and 4.2% are foreigners.

4.2. Exploratory Factor Analysis

As suggested by Costello and Osborne, EFA was used to refine and to delineate the variables of the destination attributes dimension in order to adjust this dimension by extracting the factors and by analyzing their relation [55]. To do so, this study used the oblique rotation method because oblique methods allow the factors to correlate [55]. As for the refinement, the authors suggest that it should be done by factors with factor loadings above 0.32 or that have eigenvalues with few or no cross loadings.

The following variables were excluded: AT1 (Destination is attractive), AT2 (Destination is Beautiful), AT18 (Suitable tourist signage), and AT19 (Destination has a good tourist infrastructure), 0.5 [56].

The factor structure identified three factors with responses cumulative variance of 60% in the 15 variables presented. Also, rotations converged in seven iterations for the factors’ internal arrangement.

Component 1 refers to the sub-dimension natural attractives. The dimension was adjusted with six variables, and it presented a high reliability, with a value of 0.808. The component referring to reception had the reliability of 0.796. Component 3, related to the sub-dimension services, had the reliability of 0.857. The Cronbach’s Alpha coefficients of the attributes ranged from 0.796 to 0.857, showing internal consistency of the items above the expected—0.7, lower limit [56]. The

Table 1. Exploratory factorial analysis of destination attributes.

Component
	1	2	3
AT9—Singular natural heritage	0.782
AT8—Beautiful natural heritage	0.778
AT12—Places for walking or tours	0.730
AT11—Possibilities for activities such as swimming, fishing, canoeing, and adventure activities	0.685
AT13—Beautiful beaches	0.621
AT10—Diversified natural heritage	0.521
Cronbach’s Alpha	0.808
AT5—Safe city		0.815
AT3—Welcoming destination		0.790
AT4—Welcoming and friendly population		0.788
AT6—Quality of life of the population		0.713
AT7—Modern architecture		0.433
Cronbach’s Alpha		0.796
AT15—Hotels diversity and quality			−0.832
AT16—Differentiated Gastronomy			−0.829
AT14—Good coffee shops and restaurants			−0.813
AT17—Nightlife/entertainment/leisure			−0.657
Cronbach’s Alpha			0.857

Note: AFE made with SPSS 22 software. Source: Research, 2017.

below shows the exploratory factorial analysis with refined factors. The table with all variables can be visualized in Appendix A.

It should be noted that the variables of the destination attributes come from the crossing scales process. In this process, the variables were based on Pike and Ryan and Carvalho’s scale [13,22]. From the exploratory factorial analysis, the attributes were divided into: infrastructure, accessibility, entertainment, and climate [22]. However, in this study, the sub-dimensions were configured in attributes of destination, reception, and services.

4.3. Measurement Model

The CFA is a method of data analysis that is related to structural equation modeling (SEM). This technique involves the verification of adjustments between observed data and an a priori hypothesized model, which, by its turn, is based on the hypothetical occasional relationships between latent factors (unobserved variables) and their (observable) indicator variables. In CFA, theory comes first, hence the model is a result of theory, and finally, the model is tested to achieve consistency with the data analyzed, using a SEM approach [56,57].

When verified by the Kolmogorov–Smirnov test, data presented itself as non-normal. Therefore, the robust maximum likelihood estimator (MLR) was used to analyze the data, since it is one of the most used estimators for continuous non-normal data [56]. The measurement model was estimated using the AMOS software, and the adjustment measures indicated that the data were aligned with the model.

The adjustment of the model should present indices compatible with the expected standards, that is, a root mean-square error of approximation (RMSEA) between 0.05 and 0.08.

Comparative fit index (CFI) and Tucker–Lewis index (TLI) values equaled to or exceeded 0.90 [56]. With respect to the adjustment measures presented in this study, they had adjustment results considered by Hair Jr et al. to be satisfactory and acceptable: the CFI was 0.901, the RMSEA was 0.068, the NFI was 0.84, and the TLI 0.886 [56]. Hair Jr et al. state that the CFA should provide acceptable adjustments and demonstrate the validity of the construct. The authors also point out that when the CFA values adjust and demonstrate validity, the theory can be sustained [56]. Having the theory as base, the constructs CFA, composite reliability (CR), and average variance extracted (AVE) are presented. It is worth pointing out that they will be analyzed according to the guidelines presented by Hair Jr. et al. [56].

In order to read the results, Hair Jr. et al. studies were used, because these authors point out that the standardized loading estimates should ideally exceed or be equal to 0.7, as the AVE has the same needs to be of 0.5 in order to indicate adequate convergent validity, and the CR of the construct should be 0.7 or greater to indicate adequate convergence or internal consistency [56].

As observed above in the EFA, destination attributes are configured as follows: natural attractiveness, reception and services. The analysis is based on the sub-dimensions of the destination attributes. In the sub-dimension of natural attractiveness, when reading the data, it can be inferred that, at some point, the estimate factor loadings converge with one another. As Hair Jr et al. state, standardized loading estimates should be 0.5 or more, and ideally 0.7 or more. As for the loads below 0.7, the author points out that they can be considered significant. But it is worth pointing out that they present more error variance than variance explanation, as is the case of the dimension at hand [56].

As for the CR, considering that the values suggested by Hair Jr et al. are 0.7 or more for the factor to present internal consistency, natural attractiveness presented a high index: 0.814 [56]. However, natural attractiveness did not present convergent validity, since the AVE had a value of 0.427—a value lower than 0.50, which is recommended by Hair Jr. et al. [56]. The authors state that values below 0.5 indicate more error variance than variance explanation. Thus, it is possible to point out that even if the natural attractiveness factor presents CR, the variables of the sub-dimension did not converge to validate this factor within the measurement model. In this sense, according to the data about Rio de Janeiro, the results confirm the reliability, but do not confirm the variables usefulness as precise indicators to measure the natural attributes in the studied destination. It should be stressed that replication in other destinations is important in order to be able to point out the construct validity.

The reception sub-dimension received a CR of 0.772, thus showing the internal consistency of the sub-dimension and of its variables. However, as for the AVE, it presented a result of 0.413 (though the values must be greater than 0.5). Therefore, the sub-dimension is reliable to measure the reception factor in Rio de Janeiro, but it is not valid as an accurate measurement factor for other destinations. In this sense, there may be other variables that may be included in this subconstruct.

The services sub-dimension presented a CR of 0.862, thus showing the sub-dimension and its variables internal consistency. The variance extracted presented a result of 0.609, an acceptable result according to Hair Jr et al. [56], and according to

Table 2. Confirmatory factor analysis of destination attributes sub-dimensions.

Natural Attractions	Standardized Loads	Reliability (CR)	Variance (AVE)
AT12—Places for Walking or Tours	0.727
AT13—Beautiful Beaches	0.816
AT8—Beautiful Natural Heritage	0.636
AT9—Singular Natural Heritage	0.566
AT10—Diversified Natural Heritage	0.584
AT11—Possibilities for Activities such as Swimming, Fishing, Canoeing and Adventure Activities	0.549
	3.878	0.814	0.427
Reception	Standardized Loads	Reliability (CR)	Variance (AVE)
AT3—Welcoming Destination	0.751
AT4—Welcoming and Friendly Population	0.711
AT5—Safe City	0.642
AT6—Quality of Life of the Population	0.649
AT7—Modern Architecture	0.403
	3.156	0.772	0.413
Services	Standardized Loads	Reliability (CR)	Variance (AVE)
AT14—Good Coffee Shops and Restaurants	0.813
AT15—Hotels Diversity and Quality	0.792
AT16—Differentiated Gastronomy	0.764
AT17—Nightlife/Entertainment/Leisure	0.751
	3.120	0.862	0.609

Note: CFA identification of destination attributes made with AMOS software. Source: Pesquisa direta, 2017.

, its AVE was confirmed.

Table 3. Brand equity.

Equity	Standardized Loads	Reliability (CR)	Variance (AVE)
E2—Good prices	0.906
E1—Higher quality when compared to other destinations.	0.682
E3—Leadership position	0.774
	2.362	0.833	0.628

Note: CFA identification of brand equity made with AMOS software. Source: Pesquisa direta, 2017.

indicates that the observable variables related to services are the ones that contribute the most to the composition of the destination attributes (AVE = 0.609), followed by natural attractiveness (0.427) and reception (0.413).

Regarding brand equity, it was pointed out that it had a high CR (>0.7) and a satisfactory AVE (>0.5), as can be seen in Table 2 below. The results show not only reliability, but credibility to the research instrument proposed and applied. Brand Equity is significant in both Carvalho’s and the Aaker’s thesis [13,10]. Therefore, once again it is possible to perceive the importance of brand equity theory by applying it to Rio de Janeiro, a world-famous tourist destination. By doing that, brand theory’s effectiveness can also be attested by pointing out its role on integrating elements that generate the destination branding management.

As for the brand personality dimension, observed in

Table 4. Brand personality.

Personality	Standardized Loads	Reliability (CR)	Variance (AVE)
P1—Competent destination	0.741
P2—Joyful destination	0.811
P3—Sophisticated destination	0.752
P4—Authentic place	0.748
P5—Bold destination	0.858
	391	0.888	0.614

Note: CFA identification of brand personality made with AMOS software. Source: Pesquisa direta, 2017.

below, the composite reliability was also of 0.888, presenting an excellent degree of confidence in the dimension evaluated.

As for the variance, the scale used for measurement pointed the value of 0.614, which shows that the application in Rio de Janeiro comes to collaborate with the brand personality model proposed by Aaker [15]. In this sense, it proved to be valid and reliable for this, considering that the variables used were the same in Carvalho’ thesis, and were adapted to the context of Rio de Janeiro [13]. This result serves to reaffirm the theory proposed by Aaker in 1996, since the case at hand was replicated using the tourist destination as a base, treated as a destination branding [15].

As it can be seen in

Table 5. Brand as a symbol.

Brand as a Symbol	Standardized Loads	Reliability (CR)	Variance (AVE)
BS4—Ease of Imagining the Brand	0.61
BS3—Memory and Recognition of the Symbol/Logo	0.635
BS2—Brand Features	1.288
BS1 Brand Recognition	1.000
	3.533	0.957	0.859

Note: CFA identification of Brand as a Symbol made with AMOS software. Source: Pesquisa direta, 2017.

below, brand as a symbol presented a high variable CR (0.957). Personality variables also reached an AVE greater than 0.50, as recommended by Hair Jr. et al. [56].

The results confirm the findings of Carvalho, and also corroborate Aaker’s theory, which raises the question of the symbolism importance related to brands and its importance in the identity creation in this particular case in which the identity of a tourist destination was created [13,10]. The results reaffirm and bring an important contribution to the area of tourism. They also demonstrate that in Rio de Janeiro there is a recognition of the destination symbols.

As for the brand image that, in this study, was attached to customer satisfaction, it can be seen in

Table 6. Brand image.

Brand Image	Standardized Loads	Reliability (CR)	Variance (AVE)
IM6—Resident provides a positive image of the destination	0.638
IM5—Convincing advertising material	0.753
IM4—Charming attractions	0.784
IM1—I was positively impressed	0.873
IM2—Destination stands out among other destinations	0.85
IM3—Image consistent with the way I like to see myself	0.859
	4.757	0.912	0.635

Note: CFA identification of brand image made with AMOS software. Source: Pesquisa direta, 2017.

below that it presented a CR of 0.912, and an AVE of 0.635, an acceptable value, as recommended by Hair Jr. et al. [56] (2009).

The brand image dimension is significant to compose the measurement model of this study with variables adapted from Carvalho [13]. This shows that there is a relationship between image and satisfaction which reaffirms the findings of Carvalho [13] who stated that image was identified as an important factor influencing customer satisfaction.

In the loyalty dimension, which can be seen in

Table 7. Loyalty.

Loyalty	Standardized Loads	Reliability (CR)	Variance (AVE)
L1—Recommend the destination to relatives and friends	0.928
L2—The Destination would be the first choice among those on their minds	0.817
L3—Intention to visit the Destination on the next 12 months	0.56
L4—Intention to visit the destination at the next opportunity	0.758
	3.063	0.856	0.604

Note: CFA identification of brand loyalty made with AMOS software. Source: Pesquisa direta, 2017.

, CR was also achieved (CR = 0.856). As already mentioned, this number brings confidence to the dimension variables and shows that loyalty to Rio de Janeiro can be understood through the variables L1 (I intend to recommend the destination to my family and friends), L2 (Among the different options of tourist destinations that I have in mind, Rio de Janeiro would be my first choice), L3 (I intend to visit the destination in the next 12 months), and L4 (I intend to return to the destination at the next opportunity).

Results allow us to affirm that replication evocates other studies, since the AVE presented a value of 0.604, demonstrating the convergent validity of the variables to compose the measurement model of Rio de Janeiro.

As Carvalho’s [13] thesis points out, the loyalty dimension reasserts its importance within the destination branding model. Thus, it is noticed that the loyalty to the destination, measured with the adapted variables of Carvalho [13], can be evaluated in Rio de Janeiro.

The results of the confirmatory step corroborate the findings of the exploratory phase and also show the reliability and convergent validity of some variables, which were adapted from Carvalho’s study to measure the destination branding of the tourist destination Rio de Janeiro (RJ), Brazil. The results showed that the attributes of Rio de Janeiro are appreciated by tourists through natural attractions, and services.

4.4. Structure Model and Hypothesis Testing

When testing the structural relationships of the proposed model, the results pointed to the existence of direct and significant influences. Results obtained were satisfactory, considering that they presented high indexes of adjustment, according to Hair Jr et al. In this sense, NFI presented a value of 0.991; TLI = 0.966; CFI = 0.993; and RMSEA = 0.082. These indexes show a high adjustment of the structural model, demonstrating that the relationships between the constructs can be established.

By testing the influence of destination attributes (i.e., natural attractiveness, hospitality, and services), brand equity, personality, and brand as a symbol in the image formation of Rio de Janeiro as well as the loyalty to the destination, one can perceive the following structural relationships (

Table 8. Structural relations.

Structural Relations	Standardized Loads	Errors Pattern	Loads/Standard Error	Significance
Natural Attractions → Brand Image	0.079	0.040	1.986	0.047
Reception → Brand Image	0.281	0.033	8.440	0.000
Services → Brand Image	−0.005	0.034	−0.153	0.879
Brand Equity → Brand Image	0.243	0.029	8.392	0.000
Personality → Brand Image	0.540	0.041	13.185	0.000
Brand Symbol → Brand Image	0.062	0.020	3.043	0.002
Brand Image → Loyalty	1.030	0.031	33.185	0.000

Note: CFA identification of structural relations made with AMOS software. Source: Pesquisa direta, 2017.

).

The results pointed to the existence of a significant influence (p-value < 0.05) of the natural attractions and of the reception in the brand image formation. They also confirmed the significant influence of the brand equity, the personality, and the brand as a symbol in the brand image formation. It was also confirmed the direct brand image influence in the loyalty to the destination. It should be noted that all relationships are direct and positive, and that only services are attributes of destinations that do not directly influence the formation of the brand image.

It should be noted that Carvalho’s [13] study, which was used as the basis for this study, confirmed that the set of brand associations explains the brand image. This way, his study helped to confirm that there are personality and quality relationships with the brand image; a quality impact perceived on satisfaction; a quality/equity impact perceived in satisfaction; image and satisfaction relations with loyalty; a perceived relation among quality/value and loyalty; a relation between satisfaction with loyalty; and that the brand image, the perceived quality, and the satisfaction together explain loyalty.

In this study Carvalho’s measurement model was adapted to the research universe. Thus, the use of the set of dimensions of destination branding allowed to measure the image and also its influence on loyalty.

Thus, the results of the hypotheses established a priori and tested in the study with a sample of 678 respondents indicated that:

Hypothesis H1:

The Destination Attributes influence the Destination Image. It was not confirmed, because Services did not present direct influence in the formation of the Destination Image.

Hypothesis H2:

Brand Equity influences the Destination Image formation. The hypothesis is confirmed, since Brand Equity had a direct and positive influence on the Image.

Hypothesis H3:

Brand Personality influences the Destination Image formation. The hypothesis is confirmed showing a direct and positive influence on the Destination Image.

Hypothesis H4:

Symbolism influences the Destination Image. The hypothesis is confirmed, since Symbolism had a direct and positive influence on the Destination Image.

Hypothesis H5:

Image has an influence on Loyalty. The hypothesis had a direct and positive influence on Loyalty.

The results confirm and provide theoretical and empirical support for this research hypotheses, showing that the set of destination branding variables and dimensions are reliable and precede the image formation of Rio de Janeiro. Only services did not show direct relation to the brand image formation. In addition, personality (54.1%) and reception (28.1%) were the factors that contributed the most to the brand image. This implies that, in the tourists’ perception, the image of Rio de Janeiro is seen as a set of human characteristics associated to the brand, embodying the attributes of the tourist product and adding emotional value to the experience in the destination. These results are in agreement with the studies of Keller and Aaker [3,13]. The destination image has a direct and positive relationship with the loyalty to the destination. This result is aligned with the suggestion for future research made by Kozak, Bigné, and Andreu to test and to confirm the relation of the construct loyalty to the construct of the destination image [49]. Therefore, it can be concluded that the structural model is a reliable model, and that it must be tested in other research universes in order to prove its validity.

5. Theoretical and Managerial Implications

The SEM analysis offered support for the statistically significant relationships between Brand Equity and Destination Image formation (H2), Brand Personality and Destination Image formation (H3), Symbolism and Destination Image (H4), and Image and Loyalty (H5). This study advances the destination branding literature showing that the loyalty to the destination, measured with the adapted variables of Carvalho, can be evaluated in Rio de Janeiro. Thus, the findings imply that the set of destination branding variables and dimensions are reliable measure to the destination branding of the tourist destination Rio de Janeiro (RJ), Brazil. From a practical point of view, the overall understanding of the destination branding and process of image formation regarding city of Rio de Janeiro, will help the public institutions responsible for sales management to project a suitable image of their markets by means of the best choice of communication mix.

6. Conclusions and Research limitations

The research adjusted the research instrument to the theoretical model for its application with the tourists who had visited Rio de Janeiro, thus making it possible to confirm some hypotheses put forth. It is worth noting that studies like this one collaborate with managers by giving to them a range of information that can help in the strategies linked to the brand and the image, and, potentially, the loyalty. Therefore, these studies can be used as tools of differentiation strategy thus making competitiveness and better strategic positioning possible.

As for this study theoretical contribution, we can point out the image evaluation used the features that make up the destination branding and aimed at the adjustment of the application instrument. This way, as proposed by Aaker; Carvalho; and Pike and Ryan, the study intended to get as close as possible to the reality of the studied destination [10,13,22,57]. Also, the survey research in the social network, used for data collection, allowed this study to have an acceptable number of respondents, with distinct profiles, in order to measure the proposed model.

The EFA allowed the refinement of the variables attributes and made it possible to group them into natural attributes, reception, and services. This way, more precision was provided to the CFA. Also, in this sense, it can be concluded that the observable variables related to the services are the AVE (0.609), followed by natural attractiveness (0.427) and reception (0.413).

However, in the structural model, services do not influence the formation of the destination image. The results confirmed four out of the five hypotheses established a priori in the proposed model. The features of destination branding do influence the image evaluation of the tourist destination Rio de Janeiro (RJ), Brazil. Also, personality and reception are the factors that contribute the most to a positive evaluation of the destination image. Another valuable contribution was the possibility to attest the direct and positive relationship between the tourism destination image with loyalty, a gap identified by Kozak, Bigné, and Andreu [49].

Even though this research fully achieved its aims, it presented some limitations like all scientific research. In this case, the main limitation concerned the collection of data through an electronic form disclosed through social media, which allowed a response bias. This limitation also does not allow results to be extended to the set of tourists who visit the destination. This work does not pretend to exhaust the subject; thus, we suggest further studies on the same constructs, either to replicate the findings presented here, or to overcome weaknesses. For example, the variance extracted by some sub-dimensions of the destination attributes was below the desirable minimum values (0.5), a shortcoming that does not compromise the reliability of the model but limits its applicability to the studied destination. For future research, we suggest destination branding factors to be evaluated through an alignment with other constructs, in order to prove the construct validity in the proposed measurement theory.