No One Is Leaving This Time: Social Media Fashion Brand Communities

: In an attempt to enrich existing literature on online fashion brand communities in the digital era, this research aimed at exploring the relationship between peer inﬂuence and self-disclosure on sustaining consumer engagement in generating loyalty to social media fashion brand communities (SMFBCs). The survey included a sample of 365 members who follow local Nigerian SMFBCs and was analyzed using SmartPLS v3.2.9. Findings from the study show that peer inﬂuence and self-disclosure have a positive impact on sustaining consumer engagement in social media fashion brand communities. Furthermore, the ﬁndings show that self-disclosure mediated the relationship between peer inﬂuence and sustaining consumer engagement. Finally, consumer engagement fosters loyalty to social media brand communities.


Introduction
Over the years, social media has attained commendable relevance, as it has become a ubiquitous part of daily living [1].It is estimated that over 65 million brands regard social media as a means of interacting with consumers and achieving their brand marketing objectives through social media brand communities [2].Social media brand community (SMBC), an online brand community consisting of a brand and its admirers on a social media platform [3,4], is gaining more attention as a potential means in the digital era to involve consumers in online branding activities [1,5,6].This community provides brands with the facilities to build brand awareness, have community involvement, obtain insight for better decision making, and engage in open and honest dialog [7].
With over 33.0 million employees [8], the fashion industry is among the biggest segments of the world economy, with an estimate of 3 trillion dollars, representing 2% of the global gross domestic product [8,9].This industry is one of the prominent industries that have affected every aspect of social life, causing visible changes in social [10], economic [11], political [12], climate [12], and cultural landscapes [13].The fashion industry has benefited from social media, as it has transformed the way fashion is presented, reported, and consumed [14].Top fashion events such as the London Fashion Week recorded massive success in generating "clout" on their 2014 show in which they had over half a million mentions on social media [15].In addition, fashion bloggers are given special invitations and front seats in Dolce & Gabbana fashion shows so as to enable them to upload real-time content of the show on social media [16].
The fashion industry is also among the industries that have integrated the SMBC concept into their business strategies [17,18].This can be attributed to the ability of SMBCs to enable brands to direct offline activities [19] and online traffic to their online milieu [20] so they can effectively monitor consumers' digital footprints and obtain more insight into customer journeys [21].More importantly, SMBCs provide necessary tools in sustaining consumer engagement (SCE) that generates enhanced firm performance [22].Although SMBCs are easy to create and consumers easy to recruit as community members, many brands face a lot of challenges in SCE in their communities [23].
Prior research considered consumer engagement (CE) as a vital indicator of brand marketing performance in the aspect of maintaining consumers' loyalty to the community [24,25], which is essential for the sustenance of the community.Social cognitive theory upholds this view as it posits that engagement motivates a long-term relationship that secures the loyalty of consumers to an online community [26].It costs less to maintain loyal consumers as they tend to give lesser consideration to alternatives [27][28][29].The research by [30], showed that despite the rise in popularity of online communities, there have been just a few reports of success in enhancing engagement and retaining members of the community.This report is in sync with [31], asserting that the major challenge in sustaining any online brand community is sustaining engagement and [32] highlighting that the CE phenomenon is a vital academic research agenda that has picked up the interest of brands worldwide.It is important to have loyal SMBC members as many active online brand communities have transformed into "online ghost towns" due to the inability to retain their members [33].Thus, investigating concepts of SCE in SMBCs, which subsequently enhances the community's loyalty is a critical issue in understanding the sustainability of SMBCs especially in fashion brands as there is no report to date on how to retain members of fashion brand SMBCs (SMFBCs).
Other SMBC literature exhibited that CE in these communities leads to a profitable relationship between brands and consumers [4,34,35].Some reported that the beneficial outcomes prompted CE to be enlisted amongst primacy research topics by the Marketing Science Institute in 2010 [36,37], alongside more authors calling for more insights into the dynamics of consumer engagement in online brand communities [5,32,[38][39][40][41].Despite the acceleration in the research of sustaining this concept in online brand communities, the context of SMBCs has so far been under-explored, and especially studies in the SMFBC domain are still sparse.This is an issue of concern especially for ignored African regions with emerging fashion markets such as Nigeria, ranked among the top five countries in the world experiencing the most growth in economic development, consumer spending [42], and a unique passion for keeping up with trends in the fashion business [43].Authors have suggested the need for more research on SCE in the fashion context [44][45][46], with Maria et al. (2018) highlighting that it is of importance to have more understanding on the drivers that motivate fashion consumers to keep engaging in SMFBCs and the consequences of such engagement.
In this respect, existing literature has hardly considered the effect of peer influence on generating engagement in SMBCs or SMFBCs.This is surprising as peer influence has a renowned reputation of fostering behaviors in both online and offline settings [47][48][49], coupled with shaping attitudes of individuals in the fashion realm [50][51][52][53].Peer influence applies when an individual's attitude and behavior are influenced by other members of the peer group [54].This influence can be studied using social influence theory, which shows how peers influence the behaviors of an individual in a social environment [55].Sherman et al., 2016, highlighted peer influence as a major influencer of engagement behaviors on social media.An SMFBC consists of a fashion brand and its peer consumers, and the influence of peers is of importance to online brand marketers as it encourages self-disclosure behaviors in consumers [56], which is needed for a better understanding of target consumers [57].Ref. [58], outlined that more is yet to be known on the dynamics of consumer peers on social media especially their self-disclosure behavior.
In an attempt to address the aforementioned concerns, the present research used social influence theory and social cognitive theory to enhance the current understanding of antecedents and outcomes of SCE in SMFBCs.More specifically, the study examined: 1.
The impact of peer influence and self-disclosure as antecedents of SCE in SMFBCs; 2.
The effect of SCE on loyalty to SMFBCs; 3.
Self-disclosure as a mediator between peer influence and SCE in SMFBCs.

Literature Review
The concept of engagement is one that originated from a multidisciplinary theoretical perspective such as psychology, sociology, marketing, and information systems [59,60].Various literature has provided diverse understanding of the engagement concept in relation to diverse contexts.However, it has led to a more general consensus that engagement is in three basic dimensions: cognitive, affective, and behavioral.Cognitive engagement deals with mental activities centered on something that involves attention and absorption.Affective engagement deals with the enjoyment and enthusiasm and as regards to an engagement object.Behavioral engagement comprises the active manifestations of engagement, which include sharing, learning, and endorsing behaviors [61].
Online engagement is a major factor in assessing the effectiveness of a social media platform [62].Engagement in social media is a concept exhibited by both creating and posting content [63].Ref. [64], showed engagement in SMFBCs as consumers' interactions with each other in which they share interests or reveal their perceptions on content posted by anyone with the social plugins on the platform.Ref. [65], described engagement on social media as participation and communication methods that rely on remixed texts, collaborations, and posting on social media.Ref. [66], showed that engagement on social media is functional as it involves online users' real-time interaction in the environment.
In relation to the online setting, sustaining engagement deals with constant or regular shares, comments, and the number of likes and clicks on a certain post [67][68][69][70].SMFBC platforms have the facilities to enable members to send their comments, post reviews, share experiences, like posts and comments, and interact with other community members.In other words, SMFBC platforms enable interactivity that provides many engagement opportunities.

Theories and Hypothesis Development
The current study underlined two main theories, namely social influence theory (SIT) and social cognitive theory.Social influence theory (SIT) provides a well-established foundation for understanding the behaviors of individuals in a community [71].It focuses on how an individual exhibits changes in behaviors, attitudes, and beliefs with the presence of others in the social network [72].This type of influence can be informational and normative.Informational influence occurs when one accepts information received from another as evidence about reality, while normative influence occurs when one is influenced to conform to the expectations of a group [73].Normative social influence is prevalent in communities or groups due to the desire to maintain harmony in the community or to gain positive evaluations from others [74].This is the case of self-disclosure as it plays a dominating role in maintaining harmony in relationships and influencing engagement behaviors on social media [56,75].This theory is expected to be prevalent in this study as peer influence is a type of social influence [76], which is expected to facilitate SMFBC members' compliance to keep on engaging in community activities so as to achieve a favorable reaction from community members.
Persuading consumers to stay loyal to an SMFBC is a daunting challenge due to the intense competition and reduced financial implications [77].SCT is a widely verified theory that explains the relationship existing between personal cognition, environmental influences, and behavioral outcomes [78].SCT is of the view that people have a self-belief mindset that drives them to interact and also develop loyalty to an online community [79].With SCT, consumers' loyalty to an online community is a function of their cognitive judgment on their capabilities to keep engaging in the community activities [26], which was expected to be the case in this current study.

Peer Influence
People tend to be peers with individuals that have similar features or possibly operate in the same environment [80].There is burgeoning evidence supporting that peers have a pervasive impact when it comes to engagement [81,82], with others affirming that individuals are influenced by their peers into undertaking certain actions [83,84].This was likely to be the case in this context as SIT suggests that peers in a group can influence individuals to adopt an induced behavior (engagement) as the individuals seek to get a sense of approval from the group [85].The role of peer influence has a great deal in online brand communities as consumers prefer sharing their brand-related experiences with other consumers [84].Peer actions (such as feedback) toward a member's online message play a vital role in sustaining continuous interactions [86].Online communities that exhibit peer-like relationships tend to have more engagement as peer-like relationships have stronger interactions [87].Thus, this study was of the view that: Hypothesis 1 (H1).Peer influence has a positive effect on SCE in SMFBCs.
An online brand community is a platform where one can affiliate with his or her peers.A vital aspect of affiliating with peers is self-disclosure [88].People can read personal comments or posts shared by their peers and be motivated to respond in accordance with the message [58].Self-disclosure is one of the interactive strategies that social media users adopt in interacting with their peers online [89].Individuals are usually inspired to engage in behaviors supported by their peers [90], which is a process in SIT, so as not to be seen as being out of place [74].The same goes for online brand communities as community members influence the behavioral intentions of other members by supporting them to express themselves [91,92].The trust individuals have for their peers in an online community makes them disclose information about their personal experiences [4].Therefore, this study was of the view that: Hypothesis 2 (H2).Peer influence has a positive effect on self-disclosure.

Self-Disclosure
Self-disclosure deals with individuals voluntarily and intentionally sharing their personal thoughts, feelings, and experiences [93].The nature of most SMBCs allows the creation and exchange of user-generated content, thereby facilitating highly interactive communication in the community, with opportunities for members to act as self-advocates in matters regarding their personal experience with the brand [94,95].Brands use their social media community in engaging brand-mediated intimacy in which community members disclose factual and emotional information relating to themselves [96].Some individuals also like to disclose information about themselves, as they seek to provide value to other members of the community [97,98], especially in online communities [98].Selfdisclosure is prevalent in a socially interactive setting [99], and online brand communities are relational communities that enhance social interaction among their members [100].This is similar to normative social influence when individuals tend to conform to engaging in discussion especially where discussion content presents positions favored by other community members [74].Thus this study posited that:

Hypothesis 3 (H3). Self-disclosure has a positive effect on SCE in SMFBCs.
A mediating variable is an intermediate in the causal sequence between an independent and dependent variable.It transmits the effect of its antecedent variable on the outcome variable [101].The same case was with self-disclosure in this study as shown earlier with peer influence being its antecedent and CE being the outcome variable.This shows that when an online community behaves in a peer-related manner, it generates cues that signal the extent of closeness possessed by the current community encouraging its members to keep on expressing themselves and engaging more in the community [87].This resulted in the study positing that: Hypothesis 4 (H4).Self-disclosure mediates the relationship between peer influence and SCE in SMFBCs.

Sustaining Consumer Engagement and Loyalty in SMFBCs
Studies on consumer loyalty differ, but most scholars [102,103] adopt the concept of [104], who declared that loyalty is observed when one is strongly committed to a particular product or service at the expense of alternatives.The concept of loyalty can be applied in virtual environments [70].A customer's intent of sticking with a particular SMBC as its first choice is a display of loyalty to the community.It is a positive attitude toward an online realm that involves integrating, retaining, and engaging in the virtual realm activities consistently over time [104,105].Online brand communities are platforms where community members interact and exchange brand-related information [106,107], which results in customers' loyalty to the community [108,109].Interaction on social media entails posting, liking, sharing, and commenting [110] and is an important resource of brands that drives the devotion of the community members [111].The more an individual continuously engages in an online brand community, the more the individual integrates and becomes more loyal to the community [112,113].Thus, this study hypothesized that: Hypothesis 5 (H5).SCE in SMFBCs is positively associated with loyalty to SMFBCs.

Profile of the Respondents
The study investigated key questions about associations among peer influence, selfdisclosure, consumer engagement, and loyalty.This necessitated a quantitative approach in line with similar CE studies [5,24].The study surveyed 365 members who follow local Nigerian fashion brands on Facebook.Facebook was used for this research due to its overwhelming popularity and diverse examples of active SMBCs [4,17,114].
In line with prior SMBC literature and using purposive sampling technique, the authors made sure the respondents fit into the following criteria: (1) they were members of one or more local Nigerian SMFBC [64,115].(2) The SMFBC was brand-initiated [115].(3) SMFBC page allows comments [4,116].(4) Community members were allowed to share brand-related experience on the platform [4,70].(5) There is regular release of updated brand-related information [4,116].

Measurement Scales
Items were adopted and modified from existing studies to measure the variables in the model, coupled with a five-point Likert scale containing endpoints of 1 as "strongly agree" to 5 as "strongly disagree" as shown in Appendix A.

Data Analysis and Results
To analyze the collected data, the study used partial least squares structural equation modeling (PLS-SEM), specifically SmartPLS v3.3.3.PLS-SEM is gaining vast popularity among researchers and was best suited for this study as it provides robust estimation in conceptual frameworks explaining the relationships between the variables [117] and is effective for analyzing data with a small sample size (e.g., ≤500) [118].Moreover, in line with prior SMBC and SCE literature that confirmed the applicability of the PLS-based SEM approach in predicting key target constructs in their conceptual framework [4,119,120], this study used the PLS-based approach in testing and validating the proposed hypotheses.
In determining the reliability of the survey instrument, values of the composite reliability (CR) and Cronbach's alpha (α) were used.Table 1 shows that all the CR values exceeded 0.8, which exceeds the commonly acceptable score of 0.7 [121].Therefore, the data in this study demonstrated good reliability and stability.The reliability and validity analysis conducted on the survey involved internal consistency, convergent validity, and discriminant validity tests.An array of parameters consisting of α, CR, and average variance extracted (AVE) was used in deriving the internal consistency.
As depicted in Table 1, the α of the values exceeded 0.8, which exceeds the commonly accepted score of 0.7 [121], meaning sufficient reliability.Furthermore, the analysis of the convergent validity was carried out using factor loadings (λ), AVE, and CR.As displayed in Table 1, all the AVE results were above the 0.5 threshold while CR values surpassed 0.7 and the λ exceeded 0.60, showing convergent validity and a significant level of acceptance [121,122].
The Fornell-Larcker criterion and heterotrait-monotrait ratio (HTMT) were used in evaluating the discriminant validity.The Fornell-Larcker criterion principle states that the AVE square root for each variable has to be higher than the inter-construct correlation of the research model [123].The results shown in Table 2 depict that the AVE square roots (in bold) surpass the correlation values of the variables.Based on the criteria proposed by [124], the HTMT values have to be below 1.Table 3 shows that the HTMT values are in line with the required threshold.

Common Method Bias
In order to ensure that there was no common method bias, the study employed Harman's one-factor test.Academics are of the opinion that common method bias exists if a single factor represents the bulk of covariance among the measures [4].It was observed from the results that several factors emerged and that 44.14% was the highest level of covariance explained by one factor.This shows that all the indicators could not be accounted for by a single latent variable.

Structural Model
In testing the tendered hypotheses, the path coefficient coupled with their statistical significance was analyzed using the bootstrap re-sampling function (5000 re-samples) in the Smart-PLS software.As shown in Table 4 and Figure 1, the path analysis outcome denotes that the hypothesized paths from peer influence to SCE (β = 0.210, p < 0.000) and self-disclosure (β = 0.545, p < 0.000) were significant and positive, thus endorsing H1 and H2, respectively, inferring that peer influence significantly impacts SCE and self-disclosure.Furthermore, as hypothesized, self-disclosure (β = 0.595, p < 0.000) not only showed a positive and significant effect on SCE but also positively mediated the relationship between peer influence and SCE in SMFBCs (β = 0.324, p < 0.000), thus endorsing H3 and H4 accordingly.Finally, SCE (β = 0.493, p < 0.000) showed a significantly positive effect on consumer loyalty to SMFBC.

Theoretical Contributions
This research advances knowledge in the theoretical realm of SCE and SMFBCs in particular.From the lens of SIT and SCT, the current study extends the body of knowledge on SCE by conceptualizing peer influence and self-disclosure as antecedents of SCE which leads to loyalty in SMFBC.This study also offers a pioneering investigation into the role of self-disclosure as a mediator between peer influence and SCE in SMFBCs.This answers the call for more research on SCE and SMFBCs [5,[38][39][40][41].Specifically, the following contributions were made toward the SCE, SMBC, and fashion literature: The first contribution lies in investigating the role of peer influence and self-disclosure as antecedents of SCE.This result extends the work of [51,98] by showing that peer influence and self-disclosure are influencers of consumer behavior as they drive consumers to engage in SMFBCs.
The second contribution of this study is in showing that self-disclosure mediates the relationship between peer influence and SCE.To the best of our knowledge, this study is the first to investigate this relationship thus extending consumer self-disclosure literature [58] and the importance of self-disclosure for organizational benefit.
The third contribution of this study lies in showing the effect of SCE in securing the loyalty of consumers to an SMFBC.Prior literature has investigated this relationship in various contexts [113,116], but none has streamlined the effect of this relationship to the SMFBC realm.
Lastly, this study was the first to investigate the relationships that exist among various variables in the Nigerian SMFBC context, thereby extending the SMFBC and SCE literature [18,116].

Practical Implications
The findings from this study have a number of managerial implications for managers.First, in line with H1 revealing that peer influence positively correlates with SCE in SMFBC, managers of local SMFBCs in Nigeria should pay attention to their SMFBC environment making sure that it operates in a peer-like manner where all the members (both new and old) feel equal and comfortable to make the members feel confident and interested in engaging in the community.As a result of ensuring this type of convenient setting, the community members will be more interested in engaging in the community.This can be achieved by welcoming brand-related topics, relating with all community members equally, and moderating discussions between community members in a neutral and subtle manner.
Second, H2 and H3 showed the pervasive effect of self-disclosure as it serves as both an SCE driver and a mediator between peer influence and SCE in SMFBCs.Marketers and managers of local SMFBCs in Nigeria should be keen on making sure their community members feel comfortable in disclosing their brand-related personal experiences because this will strengthen engagement in the community providing more insight to fashion brands on their consumers.
Finally, H3 showed that sustained engagement in the community leads to the loyalty of the members.This implies that managers of local SMFBCs in Nigeria should keep their social media platforms active to make sure their community members are constantly engaged.This will not only make the members focus on the community but also ensure that the fashion brand will have a sustained and reliable fan base where their products can be marketed.

Limitations and Further Research Directions
This study had a few limitations that open up avenues for further research.First, this research adopted composite measurements for the behavioral aspect of SCE variables; thus future studies can measure other aspects (cognitive and emotional).Future studies can measure other activities of the SCE aspect separately by adopting (for instance) the dimensions in [61].Secondly, results obtained from this research cover local fashion brands

Table 1 .
Measures and factor loadings.

Table 4 .
Summary of hypotheses test results.