Demystifying factors fueling university brand evangelism in the higher education sector in Tanzania: A social identity perspective

Abstract Extant literature examines the role of alumni in supporting higher education institutions (HEIs) through donation behavior. However, scant empirical studies examine the influence of alumni in the context of extra-role brand building behaviour in HEIs. This study examines university identification and university social community identification in influencing university brand evangelism when mediated with a sense of belonging. The study collected data from 606 alumni of 15 HEIs in Tanzania and analyzed it using structural equation modelling. The findings reveal that most alumni engage in university brand evangelism when they are highly identified with their university and its entire social community. Furthermore, the findings confirm that a higher level of identification is inadequate to predict and explain university brand evangelism, unless a sense of belonging is included as a mediator. Therefore, HEIs’ management should build social and personal identification to stimulate a sense of belonging, which fuels alumni intent to engage in university brand evangelism.


PUBLIC INTEREST STATEMENT
Alumni are considered crucial stakeholders who play a vital role in the growth and development of higher education institutions. At the global level, there is a significant increase in alumni support through donation behaviour. However, there have been challenges in promoting alumni to participate in supporting higher education institutions through other extra-role behaviors such as university branding. Higher education institutions can mentor their alumni and participate in supporting higher education institutions through recommendations or opinions. Following the role of social identification in boosting alumni intention to support higher education institutions, the study investigates university identification and university social community identification in creating alumni intention to support higher education institutions as brand value co-creators. This research looks into the factors that influence alumni's intentions to support their higher education institutions as brand value co-creators.

Introduction
Higher Education Institution (HEIs) are undoubtedly facing ever-growing competition and highly demanding stakeholders, i.e., students, alumni, etc. (Miotto et al., 2020;Yu et al., 2018). HEIs have been facing tremendous pressure due to internalization, ever-growing demand for innovation in teaching and learning, and constant changes in the labor market (Hemsley-Brown et al., 2016;Hemsley-Brown & Oplatka, 2015). HEIs are expected to participate in ensuring employability, knowledge and innovation transfer, and community engagement (El Nemar et al., 2020;Miotto et al., 2020). Besides, HEIs are experiencing fierce competition in the educational market when attempting to enroll the best students, recruit qualified faculty members, mobilize funds for constructing and maintaining attractive campuses, and partnering with corporate sectors (Miotto et al., 2020;Moogan, 2011). Evidence indicates that highly ranked universities in the world have been transforming their operations beyond teaching and learning to be more socially responsible and accountable (Miotto et al., 2020). However, building a world-class university is confronted with little support from potential stakeholders, including alumni, and scarcer funds due to a significant decrease in government subvention, which devastates their financial sustainability and competitiveness Miotto et al., 2020).
Evidence indicates that the higher education sector in Tanzania has not been safe from these global challenges in the higher education sector (Amani & Charles Feng, 2022a). In Tanzania, the sector is confronted with several challenges, including a decrease in state funding, which leads to budgetary challenges for the majority of HEIs (Amani, 2022b;Mgaiwa, 2018). Over the past 60 years, the country has experienced significant growth in the sector in terms of student enrolments and the mushrooming of HEIs (Amani & Charles Feng, 2022a). However, the majority of HEIs are still facing challenges in attracting qualified staff and students, corporate collaborations, research funds, etc. (Mgaiwa, 2018). Given the country's historical context, higher education was framed by socialism ideologies and thus perceived to be a typical social service and free for all (Provini, 2019). Overall, the government provided subsidies to HEIs, notably public HEIs, to carry out their routine and basic functions (Amani & Charles Feng, 2022a). However, after Structural Adjustment Programmes (SAPs) which were advocated by the World Bank (WB) in the 1990ʹs, several reforms were adopted which withdrew the government's ability to offer subsidies to HEIs (Amani, 2018). Therefore, conditions under SAPs forced HEIs to operate using a cost sharing approach. Through this approach, HEIs were supposed to operate through tuition fees and stakeholders' funds, e.g., strategic collaboration, etc. (Amani, 2018). In this case, HEIs in Tanzania had to make strategic reforms, like adopting businessoriented strategies, in order to stay alive in a commercialized and competitive higher education sector (Provini, 2019). Therefore, at the global level, including Tanzania, HEIs have been found in a dilemma, imposing pressure on their management to take deliberate strategic measures to survive in competitive settings in educational markets (Heffernan et al., 2018;Yu et al., 2018). During this time, the concept of marketization and commercialization emerged in which the higher education sector was considered part of service marketing, HEIs were viewed as service providers, and students were categorized as customers (Koskina, 2013;Lee et al., 2020). HEIs have to adopt market orientation strategies in which they are treated as distinctive corporate brands to build a competitive advantage in the educational market (Yu et al., 2018;Bagautdinova et al., 2015;Moogan, 2011). However, although HEIs adopt several marketing efforts used by for-profit organizations, including branding, scholars maintain that branding in HEIs is still in its infancy (Aspara et al., 2014). Scholars suggest that HEI branding is an inclusive process that requires the participation of various stakeholders, including alumni and students (Aspara et al., 2014). However, more focus has been on the role of alumni in supportive behaviors such as donation behavior, with little attention to university branding (Stephenson & Bell, 2014).
Several studies indicate that students or alumni with a higher level of identification are ready to support their HEIs through participating in brand-building behaviors such as word of mouth (Stephenson & Bell, 2014). Balaji et al. (2016) suggest that university brand prestige and university brand knowledge are the engines that fuel student-university identification. Stephenson and Yerger (2014) discovered that highly identified alumni exhibit supportive behaviors such as donation behavior and university advocacy. In addition, Stephenson & Yerger (2015) argue that satisfaction, when mediated with university identification, increases the overall sense of belonging to the university community and plays a role in the defining of the self through association with the university. Despite empirical studies arguing that identification plays a critical role in brand building behavior, little attention has been paid to brand evangelism (Kinyongoh, 2019;Ghani & Ibrahim, 2018). Becerra and Badrinarayanan (2013) argue that brand evangelism is the highest form of positive Word of Mouth (WOM) expressed by an individual who is an extremist and fanatic of a certain brand. While positive WOM is influenced by a short-term state of mind such as satisfaction, brand evangelism is behavior expressed by an individual who has a long-term relationship with a brand and therefore engages in preaching about the brand to recruit others to choose the brand (Doss, 2015;Marticotte et al., 2016).
This current study aims to examine the influence of university identification and university social community identification in building university brand evangelism. The study looks at Tanzania, which is in Africa, to fill policy and practical gaps about the role of HEI stakeholders, such as alumni, in building university brands. It contributes to policy issues through developing a policy framework useful for HEIs when engaging alumni in HEI branding. Using social identity theory, it is theorized that personal identity, which indicates alumni specific attributes, and social identity, which covers group-based attributes, influence alumni to behave as university brand evangelists. Hence, the contribution of the study is twofold: First, the study extends the HEIs branding concept by exploring university brand evangelism as an outcome of personal identity and social identification. Second, the study examines the influence of personal identification in building social identification, and thereafter, the study examines simultaneously the influence of personal identification and social identification in turning alumni into brand evangelists via a sense of belonging. Thus, the study has the following research question: what is the mediating effect of a sense of belonging in the relationship between university identification, university social community identification, and university brand evangelism? Through this question, both direct and indirect relationships were tested among variables of the hypothesized model of the study as presented in Tables 4 and 5.

Social identity theory
The study is founded on the social identity theory propounded by Tajfel (1974) and Tajfel and Turner (1985). Social identity theory suggests that individuals cultivate self-categorization through building enthusiasm with members of a certain social group or social community to build social identity. In this context, "social identity" refers to an individual's perception of belonging to a specific social group, social community, or social category (Markovsky et al., 1990). In addition, the social group, social community, or social category represents a group of individuals with a shared social identification or recognize themselves as potential members of the same social group, social community or social category through distinguishing themselves from non-members of a given social group or social category (Stets & Burke, 2000). Therefore, the theory is constructed on the idea of intergroup relationships, entailing the way individuals make sense of their importance as members of a particular social group (the in-group) while comparing themselves with other groups (the out-group; Hogg & Turner, 1987).
Theoretically, this form of categorization leads to ethnocentrism, which is the tendency of a person to develop a positive bias in their preference for a loving entity or object (Hogg & Turner, 1987;Stets & Burke, 2000). Often, ethnocentrism may influence the person to take extreme actions to hurt non-members and their entities or objects, becoming the worst enemies of those entities or objects. Accordingly, social identity theory recommends that individuals choose to be known or recognized in the context of their in-group membership, creating the bottom line for differentiating with outgroup members (Tajfel & Turner, 1986). Supposedly, in-group membership favoritism, as propounded by social identity theory, provides theoretical insights into the basic process in which university identification and university social community identification trigger university brand evangelism when mediated with a sense of belonging. Therefore, drawing from social identity theory, the study proposes that university brand evangelism is a form of ethnocentrism expressed by alumni who are highly identified with the university and university social community identification.

University identification (UNID)
University identification indicates the definition of self by alumni or students associated with the university brand (Balaji et al., 2016). Social identity theory suggests that individuals usually possess unique features that emanate from a certain social setup due to belonging to social groups. Nevertheless, each bears idiosyncratic characteristics that distinguish the individual from others. University identification is a category of social identification that transpires the moment students' beliefs about the university grow into a self-definition (Dajani et al., 2021;Pinna et al., 2018). It refers to students' voluntary, vigorous, and selective strong social ties with the university, with the goal of meeting one or more of their needs (El-Kassar et al., 2019). Therefore, strong social ties of between student and the university are created through this strong tie, which helps the student to accomplish their objectives at the university (Schlesinger et al., 2021). In this regard, the university comprises a structure of social identification that contributes to realizing the needs of its various groups of potential stakeholders, including needs for distinction, socialization, or self-identity (Pinna et al., 2018). Seminal work in university branding considers university identification as an attitudinal construct made up of cognitive and affective dimensions, which are theoretically considered to complement each other (Fazli-Salehi et al., 2019). Within the context of the cognitive dimension, university identification is considered the tendency of students to develop self-categorization or self-belonging to the university. On the other hand, the affective dimension suggests students' passionate or emotional commitment to or bond with the university (Schlesinger et al., 2021). Therefore, university identification is a cognitive process of overlapping or belonging between stakeholders such as alumni, students, etc, and the university. In this regard, the cognitive state influences students' or alumni affective states towards the university and their subsequent behavior towards it. Pinna et al. (2018) posit that university identification creates the link between university and students, stimulating intrinsic behavior beyond loyalty, repurchase, etc. Scholars maintain that university identification encompasses a positive emotional appraisal between the specific expectations of stakeholders such as alumni, employees, and students and overall specific expectations of the university, the result of which is a strong bond and involvement of the alumni, employee, and students with the university (Schlesinger et al., 2021). Authors such as Pinna et al. (2018) suggest that university identification symbolizes an ideal affiliation state, more extreme than merely loyalty or fidelity. Theoretically, while loyalty takes the form of perceptions that the stakeholders develop from their own experiences, university identification is related to the causes, purposes, or values the university has of which the stakeholder shares. Other scholars consider that university identification comprises a psychological state of stakeholders such as students, alumni, etc., with their respective universities, which inspire them to execute expected behaviors and extra-role behaviors to achieve the university's strategic goals (Schlesinger et al., 2021). Bagozzi and Dholakia (2006) contextualize social community as a specific group of individuals with a shared passion for a specific object or entity and well-built social identification, whose members participate jointly in various group practices to achieve specific group goals or express mutual opinions or commitments on issues affecting the interest of the social community. Evidence shows that the social community has been an important engine in determining the various behaviors of individuals due to group-based influence. Muniz & O'Guinn (2001) define a brand community as a social community within a specific and non-geographically classified social community based on a distinct set of community bonds among fans of a specific entity or object. It is widely accepted that social communities or groups function as social identifications and commonalities (McAlexander et al., 2002). Muniz & O'Guinn (2001) states that commonalities or social identification spark behavior that distinguishes key characteristics of a group or social community: consciousness of kind, rituals and traditions, and a sense of moral responsibility On the one hand, the consciousness of kind covers the intrinsic psychological state associated with the connection an individual has with members of the social group or social community (ingroup members) and the shared enthusiasm that they differ with non-members of the social group or social community (out-group members). It can therefore be said that consciousness of kind elicits a sense that a member of a certain social group or social community is to some extent different and distinct from non-members. Kuo and Feng (2013) argue that members of a social group or social community with the consciousness of kind have a shared understanding of belonging and often distinguish and occasionally abandon non-members of the social group or social community. On the other hand, rituals and traditions include sharing stories and experiences about being members of the social group or social community, reflecting on the group's history or social community, or sometimes demonstrating specific behavior to support the social group (McAlexander et al., 2002). In addition, a sense of moral responsibility consists of practices such as inspiring others to become members of the social group, developing loyalty, and sharing information with other non-members to recruit them into the social group or social community. Therefore, university community identification is a psychological state in which students or alumni develop a consciousness of self-categorization that motivates them to feel part of the university. Literature suggests that a university as a social community or social group consists of different groups such as students, faculty members or employees, parents, suppliers, alumni, government agencies, etc. (McAlexander et al., 2005). It is widely accepted that the emphasis on viewing universities as a social group or special community has gained popularity in recent times, when HEIs have been struggling to establish mechanisms to ensure student retention and reduce dropout rates. Hence, the study contextualizes that alumni develop identification towards a university when they observe a congruent between their self-identity and self-image with the university's overall corporate reputation.

Sense of belonging (SOB)
A Sense of belonging is the psychological feeling of being a family or member of a certain social group or social community (Ahn & Davis, 2020). It is an emotional state of being accepted by, contained within, and connected to a specific social community or social group (Hausmann et al., 2007). Thus, a sense of belonging provides room for socialization and fellowship between members of the social group, strengthening their social bond (Bowden, 2011). Other scholars defined a sense of belonging as a sense of membership that functions as an individual's perception of their involvement in different social situations and the support they experience from those around them (Knekta & McCartney, 2021). Literature suggests that, a sense of belonging influences a person to present themselves or prefer recognition as part of a certain social community or social group. Often, they would love to be recognized or be known as part of the community (i.e., I feel proud of belonging to this community; Knekta & McCartney, 2021). Furthermore, they would love to be connected, identified, or associated with the social group's success as it offers room for them to build and strengthen the integrity of their self or it has special meaning to their self-identity or self-image (Hausmann et al., 2007).
Therefore, a sense of belonging involves sharing values, characteristics, and ideologies among members of the social community and social group (Hausmann et al., 2007). It is an outcome of identification with the social group, as individuals feel they share the same characteristics or attributes with other members of the same social group. It influences individuals' feelings of mind that influence individuals to attach to a certain social community or social group and define themselves in relation to the social community or social group (Davis et al., 2019). It is widely accepted that a sense of belonging influences individuals to develop a social bond with a social community or group. Therefore, a sense of belonging stimulates the intention to stay or the intention to retain or hold membership in the social group or social community. Within the context of educational services, a sense of belonging has recently been considered a mechanism that motivates student intention to stay, academic performance, reduces the dropout rate, etc. (Davis et al., 2019).

University brand evangelism (UNBE)
The research stream in branding argues that brand evangelism is an advanced form of positive word of mouth (WOM; Kinyongoh, 2019). It is behavior that involves advocating for the favorite brand while disapproving of the competitors' brand through oppositional loyalty (Doss, 2015). Therefore, unlike WOM, which involves sharing positive recommendations about the brand, brand evangelism focuses on sharing, convincing, and proselytizing others to transform them into regular purchasers of the favorite brand (Becerra & Badrinarayanan, 2013). Seminal work in advanced brand-building behavior advocates that brand evangelism results from a long-term relationship between the customer and the brand. In other words, brand evangelism is relational-based behavior, while positive WOM transactional-based behavior occurs due to a shortterm state of mind, particularly satisfaction (Anggarini, 2018). Therefore, positive WOM may disappear during service failure or may change to negative WOM. On the other hand, brand evangelism is relational specific and occurs when an individual has instilled the brand values and therefore considers the brand as part of the self and represents the integrity of the self (Doss, 2015). In his commendable book titled "Selling the dream," Kawasaki (1991) contextualizes brand evangelism as an unpaid form of brand-building behavior in which a fanatic about a certain brand engages in advocating the brand to potential customers. Therefore, brand evangelists are fanatics, apostles, zealotry, etc. who participate in brandbuilding behavior whose effect goes beyond positive WOM (Hsu, 2019;Kemp et al., 2012). Scholars argue that brand evangelists are involved in brand purchase behavior, meaning that they are regular or constant buyers of the favorite brand. Therefore, they are ready to remain loyal to their favorite brand regardless of price fluctuations, service failures, etc. Often, brand evangelists prefer to defend the brand through tolerating service failure and expressing forgiveness behavior in the case of disappointment caused by the brand (Kinyongoh, 2019). On top of that, brand evangelists participate in positive referrals in favor of their favorite brand by presenting their distinctiveness and uniqueness (Marticotte et al., 2016). However, the most important behavior that distinguishes brand evangelism from other brand-building behaviors is oppositional loyalty towards competitors' brands (Amani, 2022b;Marticotte et al., 2016). Usually, this behavior focuses on challenging competitors' brands to protect the favorite brand. Although scholars debate whether this behavior should be promoted among customers, recent seminal research in customer loyalty advocates that customers develop dual behavior when developing loyalty (Amani, 2022b;Kinyongoh, 2019).

Development of hypotheses
Social identity theory suggests personal identity includes personal specific capabilities and interests, and social identity comprises social classes founded on group membership and in-group bias. Therefore, for university social community identification to prevail, there must be congruence between the university, alumni, and university members (Peruta & Helm, 2018). In this regard, a strong tie between alumni and the university fuels university social community identification. The study theorized that university identification and university social community identification can enhance a sense of belonging. In addition, previous studies indicate that a sense of belonging can enhance various extra-role behaviors, such as positive recommendations (Amani & Charles Feng, 2022a). Thus, the study proposes that a sense of belonging influences brand evangelism. With support from the above explanation, it is therefore proposed that H 1 : University identification is positively related to university social community identification.
H 2 : University identification is positively related to a sense of belonging.
H 3 : University social community identification is positively related to a sense of belonging.
H 4 : Sense of belonging is positively related to university brand purchase intention.
H 5 : Sense of belonging is positively related to university positive brand referral.
H 6 : Sense of belonging is positively related to university oppositional brand loyalty.

Conceptual model
As depicted in Figure 1, this study proposed a model that examines the relationship between university identification and university social community identification towards building brand evangelism with a sense of belonging as a mediator. The model was developed with the theoretical support of social identity theory, which advocates that a person develop self-categorization behavior by cultivating enthusiasm with other members of a given social group or social community with intention of building social identity. In theory, social identity refers to categorization in the form of belonging to a specific social group, social community, or social category (Markovsky et al., 1990). In addition, when a person develops self-categorization in a certain social group, they develop favoritism towards in-group behavior while disregarding outgroup behavior. Through selfcategorization, the person develops loyalty towards in-group behavior while demonstrating disloyalty towards out-group behavior. This study theorized that alumni who developed in-group behavior develop categorization in the form of belongingness that influences them to develop loyalty to in-group behavior and disloyalty to outgroup behavior through brand evangelism.

Participants
The population of this study was alumni of HEIs in Tanzania. The selected HEIs are located in Arusha, Kilimanjaro, Morogoro, Dodoma, Dar-es-Salaam, and Iringa. The study managed to involve alumni from 15 HEIs through an online survey. The study established the required sample size through acquiring a list of alumni and their corresponding email addresses from selected HEIs. To ensure representativeness among alumni in all selected HEIs, the study selected at least 40 alumni from each HEI while considering the size of each HEI. The final sample size in this study was 606 participants. Of the 15 HEIs, 10 were public HEIs and 5 were private HEIs. HEIs offered various non-degree and degree programmes in various specializations, including business and management, public health, engineering and mining, natural resources management, computer sciences, laws, education, political sciences, and sociology.

Data collection procedures
The questionnaire was developed in English. The questionnaire contained measurement items presented in Table 2 in which respondents were required to rate each statement in 5 points Likert scale, ranging from "5-strongly agree" to "1-strongly disagree" (Likert et al., 1934). The language used was appropriate because in Tanzania, the teaching language in higher education is English. Although the measurement items adopted in the study have been validated by previous studies, pilot testing of the instrument was conducted prior to the data collection exercise to confirm the reliability of the survey instruments. Pre-testing of the survey instruments was carried out to 111 respondents, and the Cronbach alpha coefficient for all constructs was 0.8 above the threshold of 0.7 (Santos & Reynaldo, 2013). Thus, after confirming the reliability of survey instruments, the questionnaire was sent to alumni through their email addresses. To ensure sufficient or satisfactory response rates, respondents were asked to participate in the study deliberately, and upon acceptance, the questionnaire was sent to particular alumni. Through using an online survey, data was collected in January 2021. Furthermore, regular follow-up through sending remainder emails was adopted to ensure a good response rate. In total, 650 questionnaires were sent to respondents, of which 606 were complete and valid, representing 93.2 percent of the response rate. Finally, Structural Equation Modelling (SEM) using AMOS 21 was adopted to develop and test the hypothesized model for the study.

Measurement scales
The study borrowed measures proposed and validated by previous studies in branding and university branding. All adopted measure scales have good reliability, with Cronbach alpha coefficients (α) > 0.7 (Tavakol & Dennick, 2011), Composite Reliability (CR) coefficients > 0.7 (Hair et al., 2010), and standardized factor loadings (λ) > 0.7 (Kline, 2015). However, whenever necessary, modifications, particularly rewording, were made to ensure the measure scale fit the higher education context. All variables were captured using a 5-points likert scale ranging from 5-strongly agree to 1-strongly disagree (Likert et al., 1934). The measures proposed by Balaji et al. (2016)) were used to measure university identification and university social community identification. On top of that, the sense of belonging was measured using a scale proposed and validated by Davis et al. (2019). Finally, university brand evangelism was measured using a scale from (Becerra & Badrinarayanan, 2013;Marticotte et al., 2016). Table 2 shows the results of the tests on the validity and reliability of the measure.

Common method bias test
The research methodology adopted in the study posed the likelihood of the presence of common method variance. More specifically, because all measures were self-reported, common method bias was likely to occur. In this regard, Harman's single-factor method was adopted to detect common method variance in the data as recommended by Podsakoff, Mackenzie, Lee, and Podsakoff (2003). The results showed that a fixed factor only explained 7.3% of the total 34.7% variance. Therefore, common method variance was unlikely to be an issue of concern in this study based on these results.

Evaluation of measurement model properties
To test the covariance structure of latent variables, the study examined Confirmatory Factor Analysis (CFA) using AMOS version 21. Overall findings presented in Table 2 show that factor loadings of all variables are greater than 0.5 and positive and significant, indicating that the items used were truly measures of the latent variables or that all items assisted in estimating their respective study constructs (Hair et al., 2010). The study tested the normality of data through skewness and kurtosis measurements. The results in Table 3 indicate that skewness and kurtosis values are within the threshold range of ± 1.96 (Field, 2013). Furthermore, the Cronbach Alpha Coefficient (α) was estimated to be greater than 0.7, indicating good internal consistency, and the McDonald Construct Reliability (MaxR(H)) is greater than 0.7, indicating good construct reliability (Santos & Reynaldo, 2013). Further statistics indicate that convergent validity was good because the value of Average Variance Extracted (AVE) for all specific variables was above 0.5 and the Composite Reliability (CR) Coefficient was 0.7 (Gerbing & Anderson, 1988;Henseler et al., 2014). These results suggest that the variables that ought to measure a single construct in the measurement model agree with each other. Therefore, largely the study constructs display sound measurement psychometric properties.
On the other hand, discriminant validity was checked and tested using procedures recommended by Fornell and Larcker (1981). To achieve discriminant validity, the value of inter-construct correlation between variables should be less than the value of the square root of AVE for each respective variable (Fornell & Larcker, 1981;Ab Hamid et al., 2017). The findings presented in Table  3 show that discriminant validity was achieved, and hence variables that are supposed to measure a specific construct are unrelated. In other words, the results indicate the distinctiveness of the study variables. The model fit statistics suggest an acceptable fit of the measurement model as follows: x 2 = 323.473 (p < 0.001, df = 140); x 2 /df = 2.311 slightly below the threshold of < 3 (Hooper et al., 2008;Kline, 2015). Other goodness of fit indexes indicate that, Comparative Fit Index (CFI) = 0.931, Goodness of Fit Index (GFI) = 0.953; Adjusted Goodness of Fit Index (AGFI) = 0.923, Tucker-Lewis Coefficient (TLI) = 0.972 all were above the recommended value of > 0.9 (Hair et al., 2010;Schermelleh-Engel et al., 2003), Root Mean Square Error of Approximation (RMSEA) = 0.047 below the recommended threshold of < 0.08 (Hooper et al., 2008). On top of that, Parsimony Normed Fit Index (PNFI) = 0.645 and Parsimony Comparative Fit Index (PCFI) = 0.654 above recommended threshold of 0.6 (Schermelleh-Engel et al., 2003).

Evaluation of structural model and hypotheses testing
Prior to hypotheses testing, the study evaluated the structural model goodness of fit. Overall, the structural model analysis produced an acceptable model fit as follows: To be more specific, the results of the hypothesized relationship, as shown in Table 4, show that there is a positive, significant relationship between university identification and university social community identification (β = 0.396; t > 1.96; p < 0.001). In addition, university identification has a significant and positive influence on a sense of belonging (β = 0.320; t > 1.96; p < 0.001). Furthermore, there is a significant positive relationship between university social community identification and a sense of belonging (β = 0.580; t > 1.96; p < 0.001). Finally, sense of belonging positively influences all dimensions of university brand evangelism as follows: university brand purchase intention (β = 0.273; t > 1.96; p < 0.001), university positive brand referral (β = 0.297; t > 1.96; p < 0.001) and university oppositional brand referral (β = 0.589; t > 1.96; p < 0.001).

Testing of mediation effects
The study used a process procedure for SPSS version 3.4 by Andrew Hayes (2017) to test the mediation effect of a sense of belonging in mediating the relationship between university identification and university social community identification in university brand evangelism. The findings presented in Table 5 indicate that a bootstrapping CI absolutely above zero demonstrates a significant positive indirect effect of university identification through the sense of belonging (95% C.I [0.353, 0.491], p < 0.05). In addition, a bootstrapping CI absolutely above zero demonstrates a significant positive indirect effect of university social community identification through

Discussion
The study is among the first and few endeavors in the context of emerging economies to examine both university identification and university social community identification in influencing brand evangelism. It is revealed that, university identification and social community identification play a major role in motivating alumni to engage in brand evangelism. The findings also reveal the role of a sense of belonging (i.e., I am part of this university) in mediating the relationship between university identification and university social community identification on university brand evangelism. Therefore, a positive psychological state or emotional state towards HEIs can motivate alumni to demonstrate extra role behavior as university brand evangelists. Seminal works such as Amani & Charles Feng, 2022a) support this argument by highlighting that the corporate reputation of HEIs has trickled-down effects on all potential stakeholders, including employees, students, and alumni. The present study reveals that alumni who are highly identified with HEIs develop a strong sense of belonging that eventually engenders university brand evangelism. Therefore, a sense of belonging as a result of university identification can encourage alumni to plan to buy HEIs' products in the future, make positive recommendations, and defend HEIs' brands by making negative recommendations about competitors' products.
Within the study findings, when alumni perceive that there is congruence between their selfimage or self-identity and the HEI brand, they develop a strong sense of belonging (i.e., I am part of this university). It is this sense of belonging which motivates alumni to participate in various extra-role behaviors, including brand-building behavior, donation behavior, etc. Previous studies such as Stephenson and Yerger (2014), reveal that university brand identification can stimulate alumni to build intent to support their HEIs through donation behavior. Other research domains in brand building, such as Balaji et al. (2016) investigate the role of university identification in determining brand-building behaviors such as university affiliation, advocacy intentions, suggestions for improvement, and participation in future activities. Their findings support the current findings that students who identify more strongly with their university are more likely to engage in university-supportive behaviors.
Furthermore, the study findings revealed the contribution of university social community identification towards building alumni intent to participate in university brand evangelism. These findings are supported by studies by Fujita et al. (2017) that show membership in a university social community can influence members, including staff and students, in extra role behaviors such as branding. Social identification theory suggests that in-group members express their uniqueness by comparing themselves with outgroup members, which suggests that in-group members would love to be perceived and treated as peculiar by outgroup members. Therefore, a higher level of university social community identification can motivate alumni to build a strong sense of belonging that consistently triggers university brand evangelism. Alumni use brand evangelism to show how their HEIs are different from other HEIs and to try to get people from outside their HEIs to join.
A further plausible explanation regarding the study findings is that alumni maintain or hold on to their in-group membership in the university social community through the regular purchase of HEIs offerings or by developing a tendency to participate in various social and academic events prepared by the HEIs. In addition, alumni can participate in positive recommendations about HEIs to differentiate between their chosen HEIs representing their selfimage or self-identity as a member of the university community. On the other hand, the behavior of in-group members to distinguish themselves against outgroup members motivates alumni to develop a very strong sense of belonging, which motivates them to engage in negative brand referrals to outgroup members and their respective HEIs. Drawing from social identity theory, negative brand referral is an attempt by alumni to present the superiority of their respective HEIs to the educational market. Therefore, alumni can engage in negative brand referrals against other HEIs to defend their chosen HEIs' reputational assets and their membership in the HEIs. Overall, these results back up what McAlexander et al. (2005) say, which is that social community is the most important factor that makes members of HEIs act in ways that are helpful to their own HEIs.

Conclusion
This study endeavored to investigate the important role that university social community identification and university identification play in generating a sense of belonging, which leads to brand evangelism. Having collected data from a Tanzanian-based university, the study tested the proposed framework by conducting quantitative research. The research findings confirm the positive effects of university social community identification and university identification on brand evangelism via a sense of belonging. The research confirms the pivotal role of a sense of belonging as a mediator in enhancing the relationship between university social community identification, university identification, and brand evangelism. This indicates that university social community identification and university identification are becoming continuously imperative for HEIs to build a successful corporate reputation. Based on the research findings, this study offers managerial contributions for decision makers and brand managers in HEIs who want to understand the relationship between university social community identification, university identification, and brand evangelism through a sense of belonging.

Implications
Theoretically, the study investigated the under-researched question of drivers that can trigger university brand evangelism. Within the framework of social identity theory, the study proposes the theoretical understanding that when alumni develop personal identification and social identification to hold on to their group membership in HEI communities, they develop the intent to participate in extra-role brand building behaviors such as university brand evangelism. Therefore, through this study, social identity theory is approved as adequate to explain why and how alumni engage in advanced supportive behavior towards their HEIs. In terms of practical implications, HEIs should create institutional arrangements that provide room for alumni to participate in extra-role brand building activities such as corporate reputations as brand value co-creators. HEIs should consider partnering with alumni through establishing strong alumni associations and ensure each member of the association offers maximum effort in extra role branding efforts as university brand value co-creators. It is also recommended that alumni should be invited regularly to participate actively in various academic and social events organized by HEI. Regular attendance at these events leads to social identification and a sense of belonging, which can convert alumni to be university brand evangelists. Management of HEIs should also focus on using social media to ensure alumni are consistently connected with universities. It is widely accepted that strategic tools like websites have attributes that allow alumni to be connected with members of the respective HEI.