Inglês (pdf)
Artigo em XML
Referências do artigo
Tradução automática
Enviar este artigo por email
Citado por SciELO 
Similares em
SciELO 
Permalink
INTRODUCTION
Nowadays, sports organisations are more than sporting entities, they are considered true businesses and brands that relate to each other in a sports ecosystem (Kunkel & Biscaia, 2020). In this way, several kinds of research have considered brand equity a fundamental concept for professional sports teams in different situations — e.g. male football (Biscaia et al., 2016), female football (Doyle, Kunkel, Kelly, Filo, & Cuskelly, 2021), a new sports team in baseball (Wear & Heere, 2020) and in Australian Football League (Kunkel, Doyle, Funk, Du & McDonald, 2016) — because its dimensions have associated with consequences, for instance, team identity (Wear & Heere, 2020), as well as the fan loyalty (Biscaia, Correia, Ross, Rosado & Marôco, 2013). However, the literature on this field is majority focused on North American and European realities, which is difficult to know the applicability of this knowledge in different contexts (Miranda et al., 2021).
Football teams in Brazil have a strong capacity to develop brand equity in the consumer’s mind, as this sport is considered the country’s largest social phenomenon (Guterman, 2009). Furthermore, in 2018 the Brazilian football industry generated a 0.72% impact on the Brazilian Gross Domestic Product (GDP), moving US$ 10 billion (Confederação Brasileira de Futebol, 2018). Despite this potential, understanding the impact of brand equity from the fans’ perspective has not developed as a field of investigation in the Brazilian context. The Brazilian football market differs slightly from the European market due to its competition calendar, from January to December, as well as the existence of the State Championships (Campeonatos Estaduais in Portuguese), which are only found in Brazil (Gasparetto, Barajas & Fernandez-Jardon, 2018). State competitions have been responsible for football development in Brazil and for creating local rivalries, often with teams from the same municipality. This reality has perhaps even reinforced the bond of fans with the team in the Brazilian context. This point is fundamental, given that brand equity can change with time and local (Yoo & Donthu, 2002).
In Brazil, fans also can be part of the members’ programs, where supporters pay regular amounts in exchange for benefits associated with the team. These programs are a way to increase fan loyalty, as well as an important source of revenue for the teams. Knowing that fans develop different levels of psychological connection with the teams, few researchers have analysed such differences in perception of brand equity and its impact on behavioural intentions (Biscaia et al., 2016).
Despite the great importance currently given to the study of brands (Ströbel & Germelmann, 2020), there is a gap in the knowledge regarding the discussion of this theme in the Brazilian context. The investigation of brand equity in another context can assess if a result can be generalised beyond a particular context or an isolated situation (e.g., the Brazilian fans perceive the same associations in the same way as European fans), as well as enhance the managers’ ability to make strategic decisions (Koçak, Abimbola & Özer, 2007). This is important because high brand equity can grow sales, consumer loyalty, the ability to charge price premiums, and enhanced marketing communication efficacy (Baker et al., 2022; Keller, 1993). Thus, the purpose of this study was to analyse the perception of brand equity among members and non-members of football teams and its influence on behavioural intentions. To this end, the current work will analyse whether there is a difference in the perception of brand equity between these two types of supporters and the consequent impact on behavioural intentions.
LITERATURE REVIEW
Brand equity in sport
Brand equity in sports can be considered the appropriate additional value for a product thanks to its brand. It can be strengthened owing to the strong relationship of fans with the names and symbols of the teams they support (Gladden, Milne & Sutton, 1998). Although the analysis of brand equity from a financial perspective is a source of important information, it is through the eyes of the consumer that this asset has provided relevant information on the relationship between fans and sports teams (Biscaia et al., 2013). The understanding of brand equity from the consumer perspective is based on the seminal works of Aaker (1991) and Keller (1993), which, together with the work of Berry (2000), have been widely used for the understanding of this field in the sports industry (Miranda et al., 2021). According to Aaker (1991), brand equity refers to the set of assets and liabilities linked to the different components of a brand that will add or subtract the value offered by a product or service to a company or its customers. Keller (1993) was more direct in stating that brand equity refers to the differential effect that brand knowledge causes in response to the marketing actions of the brand. The Consumer-Based Brand Equity (CBBE) was based on the associative network model to state that brand knowledge is a node of the brand in memory in which a variety of associations are linked (Keller, 1993). In this perspective, brand knowledge was conceptualised through two components: brand awareness and brand associations.
Brand awareness
Brand awareness is understood as the strength of the brand in the consumer’s mind, reflected by the consumer’s ability to identify a brand under different conditions (Keller, 1993). At CBBE, brand awareness was divided into two subcomponents: brand recall and brand recognition. It is common to observe studies of this dimension in sports sponsorship (Kończak, 2020). However, despite being widely used, brand awareness analysis in these subcomponents seems unsuitable for well-established organisations, as is the case of football teams. Alternatively, Ross, Russell, and Bang (2008) proposed to analyse brand awareness by identifying and internalising the brand because they saw that a high level of psychological commitment to the team could lead to greater awareness. In this way, some researchers have used these subcomponents in a better manner to analyse brand awareness in sports teams (Biscaia et al., 2013).
Brand associations
More widely investigated than brand awareness, brand associations are the information linked to the brand in memory and contain its meaning for the consumer (Keller, 1993). Associations are the centre of building a strong brand so that the consumer considers it strong and successful due to the perception of its strength, favorability, and uniqueness. Gladden and Funk (2002) and Ross, James and Vargas (2006) developed probably the most used scales for analysing associations in the sport. Gladden and Funk (2002) conceptualised the Team Association Model (TAM) through a literature review and subsequent focus groups with university students. The model consists of 16 dimensions, divided into attributes (success, star player, head coach, management, logo design, stadium, product delivery, tradition), benefits (escape, fan identification, peer group acceptance, nostalgia, pride in place), and attitudes (importance, knowledge, affect). Ross et al. (2006) developed the Team Brand Association Scale (TBAS), consisting of 11 dimensions: brand mark, rivalry, concessions, social interaction, team history, commitment, organisational attributes, nonplayer personnel, stadium community, team success, and team play. Even so, there are limits to the TAM in terms of conceptualisation, as it has considered motives why fans follow teams, which are not necessarily associations with specific brands. The truth is that there is no consensus on associations in sports, and several studies have either used these models to some extent (Biscaia et al., 2013; Kunkel, Funk & Lock, 2017) or sought to identify new associations (Doyle et al., 2021). Despite this, in this study, we adopted the TBAS to analyse the associations given the good psychometric evaluations presented in previous studies.
Fan club membership
Association programs are a structured reality for football teams in Europe, Brazil, and with some exceptions, in other countries. From a relational marketing perspective, all consumers have a latent relationship that can be developed between consumers and organisations (Grönroos, 1997). From the different levels of relationships, sports teams have both occasional consumers and those who pay recurring amounts to the organisation. Although both are important to sports teams, fans who pay monthly or even annual amounts are a key source of revenue for these organisations, whether they are association program members (i.e., club members) or season ticket holders. It is important to note that while season ticket holders purchase the right to attend the games for the season, members pay a regular amount to receive benefits that go beyond being present at the games, such as promotions in the purchase of products and even voting rights in club elections (Biscaia et al., 2016).
In Brazil, fan association programs are a reality that has been leveraged more strongly since 2010 and are an important source of revenue for Brazilian sports clubs. Between 2014 and 2018, clubs increased their budgets by 42% through association programs (Estadão Conteúdo, 2019). Although this turnover is influenced by the sports results, Cardoso and Silveira (2014) observed that revenues from the sale of tickets for the games are more affected by the sports’ results than the association programs. In fact, sports success is not always enough to increase the number of members of sports teams. In this perspective, considering that individuals form a psychological connection with sports teams to satisfy different needs (Funk & James, 2004), analysing the perception of brand equity between members and non-members is essential to understand the differences between them. In addition, it is indicated that individuals with greater identification with the team tend to perceive brand equity better (Ross et al., 2008). These differences have already been pointed out by Biscaia et al. (2016) in the context of football in Portugal, with members having higher perceptions than non-members. However, in Brazil, to date, investigations that analysed these perceptions have not been found.
In this sense, the current study aims to offer the vision of a new context of the perception of different types of fans about the brand of sports teams. This is essential to understand if the dimensions observed in other contexts also apply in a relevant reality such as Brazil. This information, associated with the analysis of perception between different fan groups, could provide data to aid strategic marketing campaign decisions. Therefore, the following hypotheses are presented:
METHODS
Application context
The investigation took place in the Campeonato Pernambucano 2019 edition, one of the state football championships held in Brazil from January to April. Brazil is divided into 26 states and a Federal District, which implies the occurrence of 27 state championships in the country. These tournaments emerged at the beginning of the 20th century, mainly due to the large territorial extension of Brazil and the difficulty of holding a national championship at that time. Thus, state championships were developed with the participation of teams from all over the state, from those that participate in the country’s most important division in football, the Campeonato Brasileiro Série A, to those that do not participate in any division (Gasparetto et al., 2018). The Campeonato Pernambucano — Série A1 is the first division of the competition, and the 2019 edition had the participation of ten teams distributed throughout Pernambuco. Of the ten teams, two are relegated to the second division.
SBBE adaptation and pre-test
We chose to adapt the Spectator-Based Brand Equity scale (SBBE) (Ross, 2006) to the Brazilian context, used by Biscaia et al. (2013) in the Portuguese football context. The model showed good psychometric validity (Biscaia et al., 2013; Ross et al., 2008), and its use has led to relevant results. The scale consists of 11 dimensions, divided into ten associations (brand mark, concessions, social interaction, commitment, team history, organisational attributes, team success, head coach, management, and stadium) and internalisation. Each dimension is operationalised by three items (total= 33 items) and measured on a 7-point Likert scale (1= Strongly disagree; 7= Strongly agree).
Four researchers of the Sport Management and Marketing area in Brazil adapted the content of the scale from Portuguese of Portugal to that of Brazil. Then a face-to-face pre-test of the instrument was carried out with 170 physical education undergraduate students, who participated in the study voluntarily. Before the acceptance of the students, the purpose of the investigation was explained, and they only started if they agreed with the terms. The majority were men (70%), aged between 17 and 45 (M= 22.06± 5.175). The normality and reliability of the data were analysed using absolute values of skewness (≤ 3) and kurtosis (≤ 10) (Kline, 2016) and the Composite Reliability (CR) (>.60) (Bagozzi & Yi, 1988), respectively. The factorial validity was observed assuming factor loadings >.50 (Hair, Black, Babin & Anderson, 2018). The data were analysed in Amos 26.0.
Validation of the measurement model
After the adaptation, three items were added for the analysis of the individual’s behavioural intentions, the same as those used by Biscaia et al. (2013), measured with a 7-point Likert scale, from 1 (Unlikely) to 7 (Very likely). The adapted scale was made available online, from January to April 2019, pre-COVID-19 pandemic, through the OnlinePesquisa platform. After being informed about the objectives of the investigation, the participants were required to indicate the team they support and thus proceed to respond to the 36 items. The link to the scale was shared through social network sites like Facebook and Instagram. Finally, sociodemographic questions were added (e.g., sex, age, complete education level) and whether they are members of the club they support. Thus, the sample was built for convenience.
All participants were informed about the objectives and procedures of the study; those who wished to participate freely and in an informed way filled out the questionnaire in accordance with the rules established and approved by the Ethics and Research Committee of the Universidade de Pernambuco under CAEE number: 01757118.3.0000.5192 and opinion: 3.064.330. A total of 634 respondents who voluntarily participated in the study started the questionnaire. After data cleaning, 214 responses were eliminated because they were under 18 years old or marked the same answer in ten or more items in sequence. Subsequently, 420 respondents were included in the data analysis, with a response rate of 66.25%. The sample size is above of recommendation (n> 200) (Hair et al., 2018), and we achieved the minimum ratio proposed by Hinkin (1995). As in the pre-test, most respondents were men (78.1%), aged between 18 and 70 (M= 29.56± 10.64). Regarding the education level, the majority have a high school (40.2%), followed by those with higher education (32.9%). Data on monthly family income indicated that 18.6% had an income above US$ 2,739.72 (US$ 1= R$ 3.65; 2019). In addition, 33.6% said they are fan club members of the team they support. More details may be seen in Table 1.
Table 1 Sociodemographic data and sample consumption information.
| M (SD) | n (420) | % | ||
|---|---|---|---|---|
| Age | 29.56 (10.64) | |||
| Sex | Male | 328 | 78.1 | |
| Female | 92 | 21.9 | ||
| Level of education | Elementary school | 6 | 1.4 | |
| High school | 169 | 40.2 | ||
| University education | 138 | 32.9 | ||
| Postgraduate studies | 107 | 25.5 | ||
| Monthly family income | Less than US$ 273,97 | 17 | 4.0 | |
| US$ 273,97 to US$ 547,94 | 68 | 16.2 | ||
| US$ 547,94 to US$ 821,91 | 47 | 11.2 | ||
| US$ 821,91 to US$ 1.095,89 | 62 | 14.8 | ||
| US$ 1.095,98 to US$ 1.369,86 | 50 | 11.9 | ||
| US$ 1.369,86 to US$ 2.054,79 | 52 | 12.4 | ||
| US$ 2.054,79 to US$ 2.739,72 | 46 | 10.9 | ||
| More than US$ 2.739,72 | 78 | 18.6 | ||
| Fan club membership | Yes | 141 | 33.6 | |
| No | 279 | 66.4 | ||
| Do you usually buy tickets for club games at the box office? | Yes | 244 | 58.1 | |
| No | 176 | 41.9 | ||
| Do you usually buy tickets for club games over the internet? | Yes | 161 | 38.3 | |
| No | 259 | 61.7 | ||
For the analysis of the adjustment of the measurement model, Confirmatory Factor Analysis (CFA) was utilised with the software Amos 26.0. At this moment, the considered CR cut-off point was > .70 (Fornell & Larcker, 1981). To observe the quality of the adjustment, were considered the indices: χ2 (p> .05) and its division by degrees of freedom (χ2 / df) (< 5), Goodness of Fit Index (GFI) (> .90), Comparative Fit Index (CFI) (> .90), Tucker-Lewis Index (TLI) (> .90), and Root Mean Square Error of Approximation (RMSEA) (< .10) (Marôco, 2014). The internal consistency of each dimension was analysed using composite reliability (Hair et al., 2018). Regarding the validity of the scale, the factorial validity by the standardised factor loadings was observed, as well as the convergent validity by the average variance extracted (AVE) and the discriminant validity established when the AVE of the factors was equal or higher than the square of the correlation among these factors (Fornell & Larcker, 1981; Marôco, 2014).
Multigroup analysis and structural model
In the final step, the invariance of the measurement model for members and non-members was assessed using Amos 26.0. First, the measurement model was adjusted individually for each group. In sequence, the model’s invariance between the groups was assessed by comparing the unconstrained model (measurement weights varying freely across both groups) with the constrained model (measurement weights fixed to an equal value to both groups) (Marôco, 2014). For the statistical analysis of the invariance, the CFI difference between the constrained model and the unconstrained model was observed (ΔCFI= CFIc–CFIu). The CFI measure was considered as an alternative to the Δχ2 statistic since the latter has been questioned for its sensitivity to the size and or difference of the samples, as well as for different levels of quality of adjustment (Marôco, 2014). For ΔCFI, the model’s invariance for a value equal to or less than .01 was considered (Cheung & Rensvold, 2002). After observing the model’s invariance, the latent means of the factors between the two groups were compared, and Cohen’s (1988)d was calculated to determine the effect size of the differences between the means. Finally, a multigroup structural equation analysis was applied to measure the structural model between members and non-members. After testing for model invariance in this last stage, the two groups were joined, and the structural model was calculated for the entire sample.
It is important to realise that all the procedures in this study were approved within a larger project, approved by an ethics committee of a Brazilian public university, with CAAE number 01757118.3.0000.5192 and protocol number 3.064.330.
RESULTS
SBBE adaptation and pre-test
Initially, 22 items on the scale were modified by replacing words with the same meaning. This allowed the pre-test to be conducted. After this collection, the feedback from the sample led to the modification of one more item. The data showed multivariate normality with skewness values ranging from -2.99 to .76 and kurtosis from -1.20 to 9.36. The data demonstrated internal consistency, varying between .63 (social interaction, organisational attributes) and .86 (management). Finally, all factor loadings were above .50. Thus, it was considered that the adjustments made were sufficient to continue the SBBE validation process for the Brazilian context.
Validation of the measurement model
Skewness values between -2.62 and .38 and kurtosis from -1.29 to 5.98 attested to the normality of the data, and the adjustment of the model was considered acceptable [χ2 (440)= 976.40 (p< .001), χ2 / df= 2.22, TLI= .92, CFI= .93, GFI= .87, RMSEA= .05 (95%CI .058–.076)]. The GFI showed a value below the ideal, however, good values in the other indices allowed us to observe good model adjustment (Marôco, 2014).
Only the brand mark dimension presented problems in the CR (.43), with two items demonstrating factor loadings below .50 (.30 and .48) and AVE below the required minimum (.21). Because of this, brand mark was removed from the model. After a new CFA, the model with ten dimensions showed better quality in the adjustment [χ2 (360)= 812.44 (p< .001), χ2 / df= 2.25, TLI= .93, CFI= .94, GFI= .88, RMSEA= .05 (95%CI .050–.060)]. In addition, good values for reliability, and factorial and convergent validity are shown in Table 2. The appendix presents the results of the discriminant validity for all dimensions.
Table 2 Factor loadings for each item, reliability, and convergent validity of each factor.
| Item | λ | CR | AVE |
|---|---|---|---|
| Concessions (C) | .87 | .69 | |
| There are specific foods at the arena that I like to eat | .78 | ||
| I enjoy eating at the arena | .79 | ||
| Eating and drinking at the arena are things that I like to do | .89 | ||
| Social Interaction (SI) | .84 | .64 | |
| Being a fan of the team is a good way to meet other people | .72 | ||
| I am able to see friends because of the team | .80 | ||
| The team provides a good place to see friends | .86 | ||
| Commitment (COM) | .75 | .50 | |
| Many fans regularly follow the team | .61 | ||
| The loyalty of the fans is very noticeable | .72 | ||
| Fans have followed the team for a long time | .77 | ||
| Team History (TH) | .89 | .74 | |
| The team has a history of winning | .82 | ||
| The team has a rich history | .87 | ||
| There is a successful history behind the team | .87 | ||
| Organizational Attributes (OA) | .79 | .56 | |
| The team is very loyal to its fans | .75 | ||
| The devotion to fans by the team is very obvious | .70 | ||
| The team is heartfelt to its fans | .79 | ||
| Team Success (TS) | .77 | .53 | |
| The team has high quality players | .72 | ||
| The team is a great team | .71 | ||
| The team has good performance in competitions | .74 | ||
| Head Coach (HC) | .93 | .81 | |
| The team’s head coach does a fantastic job | .85 | ||
| The team has an excellent head coach | .95 | ||
| I like the head coach of my team | .89 | ||
| Management (MAN) | .92 | .79 | |
| The management of the club is outstanding | .89 | ||
| I like the managers of my club | .92 | ||
| The managers of my club strive to improve the team | .84 | ||
| Stadium (STA) | .79 | .56 | |
| My team’s arena has ‘‘personality’’ | .63 | ||
| The architecture of my team’s arena is attractive | .73 | ||
| The arena enhances the enjoyment of going to see the games | .85 | ||
| Internalization (INT) | .90 | .75 | |
| Being a fan of (team name) is a large part of who I am | .85 | ||
| I ‘‘live and breathe’’ (team name) | .84 | ||
| I like to think that I ‘‘bleed the colors’’ of (team name) | .88 |
λ: Standardized Factor Loadings; CR: Composite Reliability; AVE: Average Variance extracted.
Multigroup analysis and structural model
The validated model showed good adjustment for members [χ2 (360)= 516.88 (p< .001), χ2 / df= 1.43, TLI= .93, CFI= .94, GFI= .81, RMSEA= .06 (95%CI .045–.066)] and non-members [χ2 (360)= 690.11 (p< .001), χ2 / df= 1.91, TLI= .92, CFI= .93, GFI= .86, RMSEA= .06 (95%CI .051–.064)]. Table 3 demonstrates the existence of internal consistency in both models, as well as factorial, convergent, and discriminant validity.
Table 3 Data on mean, factor loadings, reliability, convergent validity, and the square of correlations to indicate discriminant validity.
| Members (n= 141) | M (SD) | λ | CR | AVE | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| .70 | .68 | .54 | .74 | .62 | .54 | .79 | .82 | .54 | .69 | |||||
| 1. Concessions | 3.77 (1.73) | .76 – .96 | .88 | .70 | 1 | |||||||||
| 2. Social Interaction | 5.59 (1.31) | .80 – .84 | .86 | .68 | .13 | 1 | ||||||||
| 3. Commitment | 5.89 (1.00) | .61 – .79 | .77 | .54 | .02 | .08 | 1 | |||||||
| 4. Team History | 5.96 (1.13) | .81 – .92 | .89 | .74 | .00 | .01 | .28 | 1 | ||||||
| 5. Organizational Attributes | 4.58 (1.25) | .71 – .86 | .83 | .62 | .21 | .26 | .19 | .09 | 1 | |||||
| 6. Team Success | 4.36 (1.08) | .62 – .84 | .77 | .54 | .11 | .15 | .24 | .23 | .48 | 1 | ||||
| 7. Head Coach | 4.13 (1.47) | .86 – .93 | .92 | .79 | .00 | .00 | .03 | .01 | .05 | .21 | 1 | |||
| 8. Management | 4.43 (1.61) | .83 – .95 | .93 | .82 | .12 | .19 | .01 | .00 | .44 | .15 | .03 | 1 | ||
| 9. Stadium | 5.41 (1.32) | .54 – .88 | .77 | .54 | .22 | .25 | .09 | .00 | .47 | .18 | .00 | .30 | 1 | |
| 10. Internalization | 5.83 (1.38) | .74 – .89 | .87 | .69 | .11 | .28 | .27 | .05 | .15 | .12 | .01 | .08 | .23 | 1 |
| Non-members (n= 279) | M (SD) | λ | CR | AVE | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 |
| .67 | .61 | .50 | .72 | .54 | .52 | .82 | .75 | .56 | .75 | |||||
| 1. Concessions | 3.28 (1.62) | .79 – .85 | .86 | .67 | 1 | |||||||||
| 2. Social Interaction | 4.75 (1.51) | .69 – .87 | .82 | .61 | .26 | 1 | ||||||||
| 3. Commitment | 5.72 (1.10) | .62 – .75 | .75 | .50 | .07 | .23 | 1 | |||||||
| 4. Team History | 5.43 (1.35) | .81 – .87 | .89 | .72 | .02 | .08 | .28 | 1 | ||||||
| 5. Organizational Attributes | 4.30 (1.31) | .70 – .75 | .78 | .54 | .12 | .21 | .20 | .03 | 1 | |||||
| 6. Team Success | 4.06 (1.13) | .69 – .74 | .77 | .52 | .24 | .28 | .29 | .29 | .35 | 1 | ||||
| 7. Head Coach | 3.89 (1.48) | .84 – .96 | .93 | .82 | .07 | .09 | .05 | .00 | .20 | .20 | 1 | |||
| 8. Management | 3.72 (1.43) | .84 – .89 | .90 | .75 | .12 | .21 | .03 | .00 | .53 | .28 | .16 | 1 | ||
| 9. Stadium | 4.97 (1.45) | .65 – .85 | .79 | .56 | .19 | .46 | .22 | .09 | .37 | .28 | .09 | .32 | 1 | |
| 10. Internalization | 4.82 (1.87) | .81 – .89 | .90 | .75 | .11 | .46 | .26 | .13 | .15 | .18 | .05 | .11 | .35 | 1 |
M: Mean; SD: Standard Deviation; λ: Standardized Factor Loadings; CR: Composite Reliability; AVE: Average Variance Extracted.
The invariance of the model demonstrated a good fit for the unconstrained model [χ2 (720)= 1,207.39 (p< .001), χ2 / df= 1.67, TLI= .92, CFI= .94, GFI= .84, RMSEA= .04 (95%CI .036–.044)] and for the constrained model [χ2 (740)= 1,272.50 (p< .001), χ2 / df= 1.72, TLI= .92, CFI= .93, GFI= .83, RMSEA= .04 (95%CI .038–.045)]. The metric invariance between the groups was attested by the value ΔCFI= .01. The comparison between the unconstrained model and the model with restricted factor loadings and intercepts (means) [χ2 (770)= 1,412.11 (p< .001), χ2 / df= 1.83, TLI= .90, CFI= .92, RMSEA= .04 (95%CI .041–.048)] attested to the scalar invariance (ΔCFI= .01). Considering the group of non-members as a reference (fixed averages at zero), Table 4 indicates the significant difference between the latent averages of nine dimensions. This partially and positively confirms the H1. The positive values of the Z test indicate better perception of the members, and the higher values of Cohen’s d demonstrate the most pronounced differences between the means.
Table 4 Comparison of latent means of dimensions between members and non-members.
| Dimension | ΔLM | Z | d |
|---|---|---|---|
| Concessions | .51 | 2.93** | .29 |
| Social Interaction | .89 | 5.95** | .60 |
| Commitment | .20 | 2.11* | .16 |
| Team History | .52 | 4.19** | .43 |
| Organizational Attributes | .32 | 2.35* | .22 |
| Team Success | .30 | 2.89** | .27 |
| Head Coach | .22 | 1.41 | - |
| Management | .66 | 4.46** | .47 |
| Stadium | .49 | 2.90** | .32 |
| Internalization | .98 | 6.10** | .62 |
ΔLM: Latent mean;
*p< .05;
**p< .01; d: effect size.
Considering the H2, the causal model with the ten dimensions and behavioural intentions indicated a good fit for members [χ2 (440)= 611.31 (p< .001), χ2 / df= 1.39, TLI= .93, CFI= .94, GFI= .80, RMSEA= .05 (95%CI .042–.062)] and non-members [χ2 (440)= 815.73 (p< .001), χ2 / df= 1.85, TLI= .92, CFI= .93, GFI= .85, RMSEA= .06 (95%CI .049–.061)]. The internal consistency of behavioural intentions was achieved for members (CR= .87) and non-members (CR= .83), and factorial validity (Member= .83 – .83; Non-member= .76 – .80), convergent validity (Member: AVE= .70; Non-member: AVE= .62), and discriminant validity were confirmed in both models.
The analysis of the invariance of the causal model showed that both the unconstrained model [χ2 (880)= 1,427.51 (p< .001), χ² / df= 1.62, TLI= .92, CFI= .94, GFI= .83, RMSEA= .04 (95%CI .035–.042)] and with both factor weights and structural coefficients fixed [χ2 (912)= 1,516.41 (p< .001), χ2 / df= 1.66, TLI= .92, CFI= .93, GFI= .82, RMSEA= .04 (95%CI .036–.043)] showed good adjustment. The model was considered invariant between members and non-members (ΔCFI= 0,01). In other words, there was no difference in the impact of brand equity on behavioural intentions between members and non-members. For this reason, the causal model was analysed for fans in general (n= 420).
The analysis of the measurement model and the causal model presented identical values, demonstrating good adjustment [χ2 (440)= 939.05 (p< .001), χ2 / df= 2.13, TLI= .93, CFI= .94, GFI= .88, RMSEA= .05 (95%CI .047–.057)]. Figure 1 shows the paths of the causal model. The analysis of these trajectories revealed that the dimensions internalisation, social interaction, and management had a significant positive effect on the behavioural intentions of the fans. Finally, the dimensions of brand equity were responsible for explaining 77% of the total variance in behavioural intentions (R2= .77).
DISCUSSION
The purpose of this study was to analyse the perception of brand equity among members and non-members of football teams and its influence on behavioural intentions. Through a CFA, the data showed a change in the SBBE model because the brand mark dimension had to be excluded. This information reinforces the necessity of analysing different contexts when considering the specifications in the branding process. Referring to H1, we were able to see a better perception of the members, which indicates that consumers with a high level of relationship have a better perception of the teams’ brand. Higher levels of internalisation among members may indicate that individuals who assume team values translate this feeling into a financial attitude. The members better perceive the team as a space for social interaction. Biscaia et al. (2016) highlighted that the experience with other people is a benefit considered in the sports consumption experience. In fact, a strong relationship with the team may be related to the psychological need for social interaction (Funk & James, 2004). In addition, other associations also showed a medium effect size between differences in latent averages (Table 4).
In Brazil, some members can vote for the election of the team’s board. Thus, it is to be expected that they have the management more present in their mind. Regarding team history, this dimension was highlighted as a strong predictor of associations in basketball (Ross et al., 2008) and football (Biscaia et al., 2013). Assuming that the psychological connection with the team is greater in the members, highlighting the team success stories could be related to their desire to improve the way they are perceived (Cialdini et al., 1976). Higher member ratings for the stadium, concessions, and organisational attributes may be related to a greater presence in the games, while the small d value for fan commitment does not indicate a relevant difference.
Finally, the second consideration refers to H2, in which we were able to see that brand equity did not impact members’ behavioural intentions significantly differently compared to non-members. This indicates the importance of management considering all fans, members or not, for the strengthening of the brand (Biscaia et al., 2016). The fact that internalisation was the strongest predictor indicates the importance of teams maintaining a strong relationship with fans so that the internalisation of team values is translated into behavioural intentions. This information is also important from the perspective that maintaining involvement with sports brands can be associated with the psychological well-being of fans (Su, Du, Biscaia & Inoue, 2022). A very interesting fact is that the management dimension is also a positive predictor of behavioural intentions, considering that this dimension is under the total control of the teams. This may indicate that individuals continue to behave positively towards the brand because they believe that its management works to meet the needs of supporters (Filo, Funk & Alexandris, 2008).
The importance of teams constantly promoting socialisation spaces for social interaction is emphasised as a leisure alternative, mainly on game day. Barros Filho et al. (2021) observed that crowd experience is a factor that influences behavioural intentions in Pernambuco football, which reinforces the importance of clubs promoting this type of experience. Considering the model’s relevance, it is important to emphasise that the dimensions of brand equity explain 77% of the variance in behavioural intentions. This demonstrates the importance of considering the different dimensions presented in the model for the strategic management of Brazilian team brands.
Managerial implications
Considering the practical contributions, the fact that members better perceive different dimensions of brand equity indicates the importance of managing the relationship with the members in a qualified way so that this constantly contributes to the team’s revenue. In general, marketing departments are fundamental to developing and strengthening the brand in the fans’ minds, emphasising internalisation and social interaction provided by the team. In the case of internalisation, creating a mural with photos of the members (Biscaia et al., 2016) or delivering products to the residences of members by some athletes could strengthen the psychological connection between the team and supporter. In addition, supporting recognised social causes, for example, may strengthen the internalisation of club values not only among members but also among non-members. Furthermore, as social interaction was a strong association in the minds of members, promoting places of interaction between supporters both pre- and post-game could enhance the presence of the brand in the mind of the fan, especially for members. In the case of the three main teams in Pernambuco, this action can be enhanced thanks to the spaces existing at the organisations’ headquarters, which do not contain only the stadium, but all the facilities related to administration, amateur sports, and leisure options for the members.
Finally, as there was no difference in the impact of brand equity on behavioural intentions between members and non-members, teams also need to develop marketing campaigns that consider all supporters, regardless of their level of relationship with the team. Together with strengthening the associations mentioned above, the development of qualified management, which is perceived in such a way by supporters, directly influences consumer loyalty. This information is particularly relevant for the teams, as qualified management does not depend on the sporting result.
The limitation in the generalisation of the results should be pointed out due to the non-representativeness of the Brazilian population, meaning investigations are required in other regions. The low number of members (n= 141), compared to non-members (n= 279), requires caution when interpreting the results and information regarding the supporter’s time and level of connection with the club. Regarding the distribution of the sample by the affective team, data were collected from spectators from all ten clubs, but not in the same proportion. Future investigations should consider analysing these different aspects in the perception of brand equity. Furthermore, there was a lack of control over the existence of a concrete fan experience with the team in the investigated season because of its importance for the analysis of brand equity (Ross, 2006). Further investigations should consider the influence of other brands (e.g., athletes, leagues) on fans’ perceptions due to the teams’ relationship with other brands in the sports ecosystem (Kunkel & Biscaia, 2020).
CONCLUSIONS
Nowadays, the qualified management of the brand is a differential factor for the long-term management of any organisation. However, not all teams realise the importance of considering the brand as an asset of the organisation (Manoli, 2020). In a country that loves football like Brazil, developing a strong brand presents itself as an opportunity for teams. For this reason, the SBBE validation makes it possible to perceive the importance of connecting teams with their fans, as well as strengthening teams as a space for social interaction between supporters. All of this, together with the perception of qualified management, can offer financial returns to the teams thanks to the importance that fans give to this. The fact that the members perceive the team brand better shows the need for qualified management of the relationship with these fans as well, considering the differences presented in our study.
