Effects of brand community social responsibility: roles of collective self-esteem and altruism

Abstract

The social responsibility of brands and consumers within brand communities is a new phenomenon that can have important implications for brands, communities, and consumers. However, previous research on the relationship between brand community social responsibility (BCSR) and brands/communities is still in its infancy and the effects of BCSR are unclear. Based on the theories of social responsibility, collective self-esteem, altruism, and brand relationship, this study explores the effect of BCSR on brand loyalty behavior (i.e., purchase and recommendation intentions) to a brand community. The empirical results of partial least squares modeling reveal that BCSR is a new way of forming brand community commitment that enhances brand loyalty behavior (purchase and recommendation intentions). Moreover, collective self-esteem partially mediates the relationship between BCSR and brand community commitment, and altruism moderates the relationship between BCSR and collective self-esteem. The findings enrich the literature on social responsibility and brand community and point out the managerial implications.

Introduction

The world is moving toward a new stage in which all people assume social responsibility and create a fair society. Groups of brand admirers, called “brand communities,” have recently begun to assume social responsibility in various forms and have become a new force, undertaking acts of social responsibility. For example, members of the Harley brand community called the Harley Owners Group endowed money and materials to help impoverished students and others after an earthquake (Harley Owners Group 2019, 2020) and during the COVID-19 pandemic (Amity 2021). Jeep and BMW brand communities regularly organize and participate in tree planting and environmental protection activities (Autohome 2012, 2021). The Fandeng Book Club, a reading brand community, whose members have spontaneously held more than 100 public welfare activities throughout China, involving the donation of books, public good classes, etc. (Meipian 2018). In contrast to a company, a brand community is an informal organization-“a specialized, non-geographically bound community, based on a structured set of social relations among admirers of a brand” (Muniz and O’Guinn 2001, p. 412). The influence of these informal brand communities on when their members assume social responsibility has been little explored.

On the basis of the CSR definition of Sen and Bhattacharya (2001), we proposed a novel concept, brand community social responsibility (BCSR), which refers to economic, social and environmental matters that involve the brand community, are undertaken spontaneously by community members or are supported by the brand. BCSR is not directly or explicitly recognized by the formal reward system and can be seen as a direct extra-role behavior of community members. Although BCSR, CSR, and brand social responsibility (BSR) all exhibit societal obligations, the source, necessity, and social expectations of their societal obligations are different. CSR practice by brand/corporation is a necessary business strategy, and society or consumers have high requirements and social expectations for it (Dahlin et al. 2020). However, BCSR activities were generated by consumer spontaneity or brand support in this study. Compared with a brand/corporation, brand communities that carry out CSR activities are mostly caring and non-essential in the first place, with less commercial purpose and lower social expectations.

In the current era of contact between consumers and brands, the focal point of marketing is shifting from “transaction” to “relationship.” Commitment is regarded as a key factor in relationship marketing and essential for an organization seeking to develop long-term relationships with customers and to ensure sustainable competitiveness (Hur et al. 2011). Brand community commitment is considered the focus of the transformation of the consumer–brand relationship (Wang et al. 2019; Zhou et al. 2012), which is a central determinant of consumer behavior intention, brand loyalty and commitment (Wu and Gong 2021). Studies have indicated that CSR can affect organizational commitment (Kim et al. 2021). Whether BCSR affects brand community commitment has not been explored. One of the goals of brand communities is to nurture and increase brand loyalty among their members, and influence consumer brand loyalty perceptions and behaviors (Yang et al. 2016). Evidence of the effect of BCSR on brand community commitment would reveal another approach to advancing brand loyalty behavior (i.e., purchase and recommendation intentions) through brand communities.

In addition, mediating and moderating effects may exist in the influence of BCSR on community commitment. Although only some members may engage in social responsibility activities, pictures and texts regarding related activities are shared within a brand community and influence other members. Consequently, these members are proud of their membership in the brand community. We employed the social psychology concept of “collective self-esteem” (Crocker and Luhtanen 1990) to describe the psychological satisfaction and pride of members in their brand community. Whether collective self-esteem in a brand community mediates the relationship between BCSR and community commitment requires examination. Another question is whether altruism moderates the relationship between BCSR and collective self-esteem in a brand community. Altruism is helping or sharing behavior that promotes the welfare of others without conscious regard for personal self-interest (Smith et al. 2006). An altruist is inclined to identify with the social responsibility activities undertaken by others and perceive collective self-esteem. Therefore, altruism may strengthen the relationship between BCSR and collective self-esteem in a brand community.

Studies have not yet explored the aforementioned, which together represent the knowledge gap that we aimed to fill. This study makes three contributions. First, it proposes the concept of BCSR, discusses social responsibility from the perspective of brand community, and widens the research field of social responsibility. Second, it indicates that BCSR is a new approach to forming brand community commitment, which can promote brand loyalty behavior (purchase and recommendation intentions). Third, brand community collective self-esteem was determined to partially positively mediate the relationship between BCSR and brand community commitment, and altruism positively moderated the relationship between BCSR and collective self-esteem. These findings may enrich the theoretical research on social responsibility, commitment, collective self-esteem of brand communities, and altruism. They may also provide practical guidance regarding how brand loyalty behavior can be enhanced through social responsibility.

The paper is organized as follows. First, based on theories of social responsibility, collective self-esteem, altruism, and brand relationship, we review the theoretical background and develop a conceptual model and research hypotheses. Second, we describe the data collection process and develop the measurements. Next, we examine the hypotheses through regression analysis. Finally, we discuss the theoretical and practical implications of the findings and indicate the limitations and future research directions.

Theoretical background

Corporate social responsibility, brand social responsibility, and brand community social responsibility

CSR represents the obligations of an enterprise to society as well as the scope of the enterprise’s activities and state related to its perceived social or stakeholder obligations (Grohmann and Bodur 2015). CSR can improve operation efficiency, reduce costs and thus help a company obtain a competitive advantage; it is also a part of an enterprise’s brand–building efforts (Dahlin et al. 2020). By participating in CSR activities, brands hope to encourage a more desirable attitude among stakeholders, and to establish a strong, positive consumer–brand–enterprise–stakeholder relationship in which an emotional bond is formed to encourage supportive behavior (Du and Sen 2010). To do so, a common cognition of CSR motivation must be constructed. Scholars have extensively explored consumer cognition of CSR, and progress has recently been made in research on consumer cognition of BSR. BSR can be defined as “consumers’ perceptions of the extent to which a brand reflects the human values (i.e. fairness, humaneness, caring, and compassion it represents) related to social responsibility” (Grohmann and Bodur 2015). BSR appears to be positively related to brand achievements. Madrigal and Boush (2008) emphasized the importance of exploring social responsibility at the brand level, and they found that BSR cognition positively affects consumer attitude toward advertising brands. He et al. (2016) confirmed that BSR image positively moderated the relationship between consumer moral identity centrality and intention to purchase cause-related marketing sponsor brands. Moreover, BSR also has positive impacts on brand loyalty (Shin et al. 2022), purchase intention (Du et al. 2007), and brand equity (Hsu 2012). The effective communication of BSR positively affects the response to the brand of consumers and stakeholders (Wang et al. 2016). This is largely because brand CSR represents the self-transcendental value of caring for society (Torelli et al. 2012). In this case, BSR is an effective and differentiated competitive positioning strategy (Du et al. 2007).

A brand community is a collection of social relationships with a brand and consumers as the core and takes into account products and stakeholders (McAlexander et al. 2002). As a unique form of relationship marketing, a brand community can have a strong impact on CSR activities through strong emotional bonds between consumers and the brand. Moral responsibility refers to the feeling of responsibility or obligation among brand community members, to the whole society, and to individual members and it stimulates collective action, thereby enhancing group cohesion and identification (Muniz and O’Guinn 2001). Community members often spontaneously share information, donate, and volunteer to implement social responsibility to ameliorate social problems. These acts increase the common sense of community and brand loyalty and create a sense of moral responsibility among consumers (Chaudhry and Krishnan 2007). However, few studies have investigated how community members initiate social responsibility. Scholars have explored whether CSR is applicable to brand communities and its role within them. For example, Hoeffler and Keller (2002) observed that creating a sense of brand community was a means of establishing brand equity through corporate social marketing for enterprises. Luedicke (2006) determined that members of the Hummer brand community followed with interest Hummer’s social responsibility plans to clearly explain the brand to individuals outside the community. A brand community is a business community (Muniz and O’Guinn 2001) with a scope that has been gradually modified and expanded. Many nonprofit organizations (e.g., the American Heart Association, food banks, and universities) build a brand community around nonprofit aims and charity to implement social responsibility (McAlexander and Koening 2012). Thus, members of the brand community created by a nonprofit organization share similar ideas and feel a responsibility to construct a more equitable society. Scholars have conducted exploratory research in this area, and their findings provide a basis for the current study. Hassay and Peloza (2009) published a conceptual model of the development of a charity brand community and proposed that cognitive and experiential learning are unique mechanisms for the development of such a community. De la Peña et al. (2018) called this type of charitable or public welfare brand community—that formed around a well-known nonprofit brand providing services—a “cause-brand community.” McAlexander and Koening (2012) explored how the brand community construct is adapted to the charitable intention of nonprofit organizations to aid in the cultivation process. They revealed that the brand community is a meaningful contributor to the expression of philanthropic intention in universities of different sizes.

CSR activities have many positive effects, which are closely related to the brand, company, community relationship, and experience of consumers. Nowadays, CSR programs are ubiquitous in enterprises, affecting stakeholders and society as a whole. Relationships among consumers, enterprises, third-party stakeholder groups, brands, and products are strengthening creating consumer groups composed of consumers and brands. Thus, the research paradigm for social responsibility has changed, from enterprise to brand and now to the brand community. According to a review of the literature, most of the social responsibility literature has been conducted at the corporate (CSR) or brand (BSR) level. However, it was found that the relationship between BCSR and brand/community has rarely been explored, and the consequences of BCSR are unclear.

Brand community collective self-esteem

Self-esteem can be personal or collective. Personal self-esteem emphasizes individualistic aspects of the self—an individual’s sense of value and respect (Brewer and Weber 1994)—whereas collective self-esteem is a collective sense of value and respect. In a collective, people regard themselves as part of a group and are strongly attached to other group members (Luhtanen and Crocker 1992). Based on social identity theory, Crocker and Luhtanen (1990) proposed the concept of collective self-esteem, which means that individuals evaluate themselves in relation to their social or cultural group membership. Luhtanen and Crocker (1992) argued that collective self-esteem is a higher-order construct that consists of four subdimensions—private collective self-esteem, public collective self-esteem, membership esteem, and importance to identity. The collective self-esteem emphasizes the sense of self-worth, satisfaction, attachment, and respect that individuals feel in their group, reflecting the difference between “ours” and “theirs” (Shang and Sargeant 2016). The use of “ours” and “theirs” reflects the group to which a person belongs. Collective self-esteem may be a decisive individual feature that can predict the attitude of group members toward their group (Brewer and Weber 1994). Individuals are eager to join high-status out groups to improve their collective identity and collective predicts the degree of preference for high-status out groups (Li et al. 2021). The collective-self is based on moral or social ability (Soral and Kofta 2020), which was also found to have a strong relationship with citizenship behavior (Roney and Soicher 2022), individuals with higher collective self-esteem resist making unethical decisions (Dissanayake 2021). In addition, collective self-esteem may be as crucial to people’s lives as personal self-esteem and may substantially affect psychological well-being (Bettencourt and Dorr 1997). Previous studies have shown that collective self-esteem can also reduce perceived stress through perceived social support, anxiety (Chen et al. 2021) and depression (Sharma and Agarwala 2013).

Brand community identification has recently been the focus of studies (e.g., Confente and Kucharska 2021). Rooted in organizational identification theory, “brand community identification” means that members who identify with a brand community regard themselves as part of the community and engender consciousness of kind (Algesheimer et al. 2005). Consciousness of kind, which is understood as “the intrinsic connection that members feel toward one another, and the collective sense of difference from others not in the community” (Muniz and O’Guinn 2001), is a key component of brand community. Collective self is made up of characteristics of the self-concept that distinguishes in-group individuals from relevant out-group members (Zhou et al 2021), and it is built on impersonal ties to others formed from common (and frequently symbolic) group identifications (Dissanayake 2021). Furthermore, the reputation of a community or achievements of other community members can induce a sense of pride. For example, CSR can improve the organizational image and reputation (Vo et al. 2019). When community members see other members participating in CSR activities, even if they do not participate, they will be proud of their membership in the brand community. Since it brings a sense of social identity and improves collective self-esteem. Therefore, members can attribute the origin of collective self-esteem to a brand community. In addition, collective self-esteem is also reflected in the “good feeling” and degree of communication within the brand community (Shang and Sargeant 2016). In brand communities (e.g., Samsung, Apple, Huawei, Halley, and Hummer), if someone maliciously attacks the community, members protect it; they attempt to maintain the collective self-esteem derived from the brand community. A key moderator of intragroup bias and service attribution, collective self-esteem can strengthen (or weaken) member attachment and commitment to social groups (Luhtanen and Crocker 1992). For example, collective self-esteem is considered the leading factor in the customer–brand relationship structure, as indicated by the commitment and word-of-mouth (Shin et al. 2022).

Brand community commitment

Commitment can be defined as the “enduring desire to maintain a valued relationship,” and it consists of emotional, continuous, and normative commitment (Allen and Meyer 1990). Organizational commitment is generated when members become familiar or satisfied with an organization, develop attachment, loyalty, identity and pride, and establish relationships (Obaze et al. 2021). The concept of brand community commitment originates from organizational commitment and refers to the desire of members to maintain relationships formed in the community (Zhou et al. 2012) or retain the attitude of members toward the community (Zhang et al. 2015). Therefore, brand community commitment can be indicated by the strength of the relationship or attitude between members of the community.

In the marketing literature, scholars have generally explored the factors influencing brand community commitment—such as brand community integration (McAlexander et al. 2002), community identity and brand relationship quality (Xiong et al. 2018), brand community trust (Hur et al. 2011), and brand community experience (Wang et al. 2019)—from the perspective of the consumer–brand relationship. Brand community commitment is the result of a two-way dynamic construction, and scholars have studied it as being constructed by organizations and consumers. Consumers need to feel close to maintain their commitment to the organization, brand, and community (Muniz and O’Guinn 2001); thus, brand community marketing managers must know how to provide conditions that encourage members to establish emotional ties with the community. The existing literature has also explored community commitment from the perspective of community management. Zhang et al. (2015) suggested that information quality, perceived expertise, and social interaction positively affect brand community commitment. Yang et al. (2016) concluded that perceived community support positively impacts community commitment. Generally, consumers join a brand community to obtain the information, entertainment, experience, and social value contained in the community. Shen et al. (2018) explored the nonlinear effects of the person-environment fit of needs-supplies fit and demands-abilities fit on community commitment, and the results indicated that needs-supplies fit had a decreasing incremental effect, while demands-abilities fit had an incremental effect on community commitment. The members of brand communities have stronger commitment than members of most other types of communities (Muniz and O’Guinn 2001), probably because they obtain various types of value and resources and establish a stable, long-term relationship. Studies based on the perspective of consumers have shown that hedonic, social, and learning benefits (Kuo and Feng 2013), perceived functional usefulness, system quality, and pleasure (Gupta and Kim 2007), and social benefits (Jin et al. 2010) have positive impacts on the formation and development of community commitment. When members benefit from participating in their community, they are willing to maintain a relationship with it, which improves brand community commitment and is conducive to marketing within brand community relationships. Therefore, brand community commitment can be understood as a positive emotional bond or attitude formed by perceiving an accumulation of community interests. It comprehensively acts on members and represents the willingness of members to maintain the relationship.

Brand purchase and recommendation intentions

Brand communities are the latest evolutionary step in supporting marketing activities in the digital network era, and strong brand communities can bring significant economic and social value to companies by increasing consumer loyalty to their brands (McAlexander et al. 2002). Repeat purchase and recommendation intentions are the two consumers’ behaviors that reflect brand loyalty (Chung and Kim 2020). Brand purchase and recommendation intentions refer to the extent to which consumers purchase and recommend others to use the brand’s products or services. Existing research on the antecedents of purchase intentions and recommendation intentions is concentrated on two aspects: emotional relationship bond and attitude, and perceived value and satisfaction.

Emotional relationship bonds and consumer attitudes are often used to explain both types of brand loyalty behavior. On one hand, brand community relationships can translate into marketing effects such as purchase intentions and word-of-mouth (Demiray and Burnaz 2019). When relationship quality is high, consumers are more likely to recommend a specific brand’s products to others and to buy more products from that brand (Huntley 2006). Brand communities reflect the relationship between a brand and its community members (Muniz and O’Guinn 2001), and highly engaged emotional relationships will influence members’ behavioral intentions toward the brand (France et al. 2016). Another thing, trust also is an important driver of purchase and recommendation intentions (Furner et al. 2022). CSR activities directly and indirectly influence consumer purchase intentions through trust (Zhang and Ahmad 2021). For example, consumers’ perceptions of giving and purchase intentions are stronger when the donation amount is larger (Human and Terblanche 2012).

Perceived value and consumer satisfaction are closely related to repurchase and recommendation intentions. Perceived value is also one of the main indicators of purchase intentions (Leroi-Werelds et al. 2014). Online brand community values (perceived and social) can have a positive impact on members’ intention to recommend a brand through community engagement (Soelaeman et al. 2022). Brand community members gain value from interactions with each other and build trust and loyalty to the focal brand (Laroche et al. 2013).

Extant research also indicated that one of the key motivations for members to visit or join a brand community is to seek the desired value satisfaction (Mathwick et al. 2008). When satisfaction is above the repurchase or recommendation threshold, customer is willing to have repurchase and recommendation intentions (Ying and Meng 2009), but satisfaction does not have the similar effect on repurchase and recommendation intentions (Soderlund 2006). Zeithaml (1988) proposed a theory of consumer perceived value from a consumer psychology perspective. Through empirical research, Zeithaml (1988) suggested that when the perceived benefit of a product or service is higher, the perception of value will be higher; and the perceived value of the product will increase the consumer’s willingness to purchase the product. Online self-presentation behavior is associated with purchase intentions in virtual environments, and perceived value reinforces the transmission mechanism between online self-image expression and purchase intentions (Chen and Chen 2020). Meanwhile, high perceived value positively impacts consumers’ purchase intentions toward brands (Sharma et al. 2021). In addition, the perceived effect and brand immersion of social media digital can increase consumers’ expected willingness to recommend a brand and their intention to purchase (Jiménez-Castillo and Sánchez-Fernández 2019).

Hypotheses

We constructed a conceptual model (Fig. 1) on the basis of the logical relationships among cognition, affection, and conation. As the model shows, when members realize that the brand community assumes social responsibility (cognition), they are proud of the brand community attracting social praise (affection) because they are a member; consequently, they remain in the community (conation). Additionally, they form a positive attitude toward the brand community when it assumes social responsibilities (cognition), and this attitude might drive them to maintain a long-term relationship with the community (conation). Therefore, collective self-esteem within a brand community may play a mediating role between BCSR and brand community commitment. In addition, not all members identify with BCSR. Altruism theory is used to explain the contribution to a charity such as charitable giving (Hill 2016), cause Marketing (Woo et al. 2020), and other sustainable or prosocial behaviors (Booysen et al. 2021); besides, altruism is also a dimension of citizenship in online brand communities, and the social responsibility activities undertaken by the brand community are also a form of citizenship. Therefore, it matches the situation and theme of our research. Altruism is particularly important in the social media environment for brand communities, where members often come and go for no reason and without monetary reward. And connections need to be strengthened through altruism by fostering loyalty, interdependence and commitment to long-term prosperity so that they make regular social contact and want to stay in the network. Although research on prosocial behavior and altruism has been ongoing, there is still a lack of consensus in predicting altruistic personality or behavior (Dargan and Schermer 2022). Thus, it is of great practical significance to study it further. Moreover, altruism is important for two reasons at a psychological level. Altruism is considered an important personality trait that helps people to be socialized (Fatfouta et al 2021). Non-altruistic members might have weaker psychological responses to BCSR than those who are highly altruistic. Therefore, altruism may have moderating effects on the relationships between BCSR and brand community collective self-esteem and between BCSR and brand community commitment. Based on the above five reasons, this study takes altruism as the moderator.

Fig. 1.

Conceptual model*. Notes:* The model does not show the hypothesis regarding the mediating effect (H4)

We selected brand prestige, community age, community size, membership tenure (i.e., length of brand community membership), forum categories, and whether to attend public welfare events as the control variables. Collective self-esteem and commitment derive from the brand community; thus, the size (number of members) and age (length of development history) of a brand community may influence members’ attitude and trust toward the community and further affect brand community collective self-esteem and commitment. Brand prestige refers to the market evaluation of a brand received after spreading (Bhat and Bowonder 2001). When a brand has a favorable social reputation (e.g., through word-of-mouth), the collective self-esteem of brand community members is maintained or strengthened on the basis of others’ evaluations of the brand (Gao et al. 2018). When a brand’s prestige drops, the brand community may be threatened. We also employed demographic characteristics such as sex, age, income, and education as control variables.

Direct influence of BCSR on brand community commitment

Enterprise practices social responsibility by organizing activities and voluntarily undertaking economic, social, and environmental responsibilities. As mentioned, studies have referred to the influence of CSR on brands. CSR image or reputation affects brand prestige and attractiveness (Hsu 2012) as well as brand loyalty (Shin et al. 2022). Two studies have suggested that CSR has a positive effect on the brand community (Chaudhry and Krishnan 2007). When a brand community assumes social responsibilities, members form a positive impression of the community, which enhances their trust in it (Park et al. 2017). This trust and positive impression may promote commitment to the brand community (Hur et al. 2011). In addition, when BCSR activities stimulate mutual trust, members are more connected and, in turn, more likely to form an emotional attachment and long-term relationship with the brand community (Dunn et al. 2015). Thus, we propose the following hypothesis:

H1

BCSR positively influences brand community commitment.

Mediating effects of brand community collective self-esteem

Because CSR has positive effects on corporate reputation (Hur et al. 2014), enterprises regard social responsibility initiatives as crucial instruments for establishing a corporate image (Vo et al. 2019). Moreover, the enhancement of corporate image through CSR initiatives increases the degree to which employees identify with the company (Kim et al. 2010). A brand community that assumes social responsibility gains public praise, which can strengthen membership and fosters brand community identity (Currás-Pérez et al. 2009) in addition to maintaining and improving self-image (Escalas and Bettman 2005). This self-image promotion through brand community engenders pride in members and collective self-esteem. Morality is the linchpin of group behavior coordination (Ellemers and Van den Bos 2012) and is often considered a key predictor of collective self-esteem (Soral and Kofta 2020). Therefore, when a brand community practices social responsibility and moral behavior (e.g., donations and environmental stewardship), collective self-esteem will be improved. Thus, we propose the following hypothesis:

H2

BCSR positively influences collective self-esteem in a brand community.

Collective self-esteem is considered a large element of the customer–brand relationship structure (Shin et al. 2022). Evaluations of a brand community increase positive affection for the community and enhance the conation of community commitment (Hur et al. 2011). Positive public evaluation of a brand community fosters pride in members and strengthens their membership (Van and Pierce 2004). Therefore, members want to maintain their membership (i.e., community commitment) to gain honor. Identification with their status and contribution is positively correlated with the perception that they are crucial to the community. The importance of identifying with collective self-esteem implies that members assess the degree to which people think highly of their membership. Thus, membership fosters long-term organizational commitment (Shin et al. 2022) because community membership advances members’ self-esteem (Mael and Ashforth 1992). Brand community collective self-esteem is the value that the community offers its members, and studies have demonstrated that community value is a crucial antecedent to brand community commitment (Mathwick et al. 2008). Thus, we hypothesize the following:

H3

Brand community collective self-esteem positively influences brand community commitment.

Combining H2 and H3, brand community collective self-esteem may be a mediator between BCSR and brand community commitment. A brand community as an informal organization assumes social responsibilities and gains societal recognition and praise. From the community identity perspective, members are proud of their membership, which induces collective self-esteem. Pride derives from the image of a brand community that undertakes social responsibility activities. The benefits of membership are collective self-esteem and gratification. Hence, members tend to remain in the community, ensuring brand community commitment. Thus, we propose the following hypothesis:

H4

Brand community collective self-esteem plays a mediating role between BCSR and brand community commitment.

Moderating effect of altruism

Human nature involves the moral attributes of self-interest and altruism, and the criterion indicating whether an action is moral is whether the action benefits or fulfills the majority of people (Marmarosh and Corazzini 1997). Altruism can be defined as sacrificing personal interests by not seeking or obtaining external benefits and by actively and voluntarily engaging in behavior valuable to others (Smith et al. 2006). Altruism is related to a sense of identity and responsibility within an organization (Yang et al. 2020). When a brand community engages in social responsibility, altruistic members are more active and willing to share and contribute, actions that will be praised by others (including the community and society) and give these members senses of self-worth, respect, and identity, further integrating them into the brand community. In addition, altruistic members are simultaneously proud of being a brand community member and willing to connect personal self-esteem to community self-esteem (Van and Pierce 2004). When a brand community assumes social responsibility and obtains a positive social image or reputation, altruistic members are more proudful or satisfied, and more willing to recognize their membership than individualistic members. Altruism increases the effect of social responsibility on collective self-esteem. Therefore, we propose the following hypothesis:

H5a

Altruism has a strong impact on the relationship between BCSR and brand community collective self-esteem.

Because altruistic members enjoy sharing and helping others, they recognize the contributions of other members to social responsibility, evaluate the community appropriately, and are likely to remain in the community to enjoy the friendly atmosphere and thus establish long-term relations with other members (Hur et al. 2011). By contrast, members who are not altruistic do not enjoy helping others and do not consider BCSR initiatives necessary. They join the community to receive rather than to give. They may believe that help from the community makes a limited contribution to society. Consequently, the social responsibility behavior of other members does not positively affect their evaluation of the community or strengthen their commitment to it. Altruism assigns social responsibility to community identity (Nan and Heo 2007). Research has suggested that community identity is positively correlated with community commitment (Obaze et al. 2021). Therefore, we infer that altruism is the primary factor in enhancing commitment to the community. Thus, we propose the following hypothesis:

H5b

Altruism has a strong impact on the relationship between BCSR and brand community commitment.

Direct effects of brand community commitment on brand purchase and recommendation intentions

A brand community is a collection of social network relationships based on brand admirers (Muniz and O’Guinn 2001). When community members are passionate about a brand, they develop brand loyalty behavior to support their preferred brand, such as repurchasing the brand or recommending it to others (Ying and Meng 2009). Previous studies have shown that brand community members’ commitment to their community has a positive impact on brand loyalty behavior (Raïes et al. 2015). Therefore, brand community commitment is likely to positively influence brand purchase and recommendation intentions. Furthermore, brand community commitment is a positive community relationship (Zhou et al. 2012), which can help to increase brand attitudes and brand loyalty behavior (Zhou et al. 2022). This means that the relationship with the brand community is better, and the community members are more willing to spend time to recommend the good aspects of the brand to others (Royo-Vela and Casamassima 2011) and to make re-purchases (Hur et al. 2011). Therefore, we propose the following hypothesis:

H6

Brand community commitment positively influences brand purchase intentions.

H7

Brand community commitment positively influences brand recommendation intentions.

Method

Sample and data collection

We selected online automotive brand forums as the setting for observing the BCSR phenomenon and collecting data. We chose this setting because (a) car owners comprise the upper-middle social class in China and are likely to donate to public welfare causes; (b) they own cars and can easily visit public welfare locations (e.g., endowed schools); (c) they often post information about a brand community on their cars (e.g., the flag or slogan of the brand community) to promote understanding of the commonwealth behavior of the community; and (d) brand community researchers often use car clubs as a research setting (Algesheimer et al. 2005).

We posted our questionnaire on “Wenjuanxing” (www.wjx.com), a popular Chinese online survey platform that maintains connections with many online brand communities, and randomly published the questionnaire web link on several active car forums centered on China. A total of 546 individuals responded over 13 days. Because of the completion time or irrationality of the responses (e.g., completing the questionnaire by using only one number), 7 questionnaires were deemed unsuitable for analysis and were deleted. The remaining 539 questionnaires were used (valid response rate = 98.72%).

These participants included male (45.6%) and female (54.4%) within the following age groups: younger than 20 years old (0.6%), 21–30 years old (56%), 31–40 years old (39.6%), 41–50 years old (3.7%), and above 50 years old (0.7%). Most respondents were in-service employees (96.1%) and had college or bachelor’s degree (99.3), and 83% had high monthly incomes (> 6000 Yuan). All the data were collected by investigating the actual OBC registered members of more than 50 different car brand forums (e.g., Mychery, Toyota, Jeep, BMW, Audi, Mercedes-Benz, Volkswagen, Ford, Chevrolet, Lexus, Volvo, Mazda, Renault, Hyundai, Chrysler, Porsche, BYD, GEELY, Chery, etc), most of which have more than tens of thousands of registered members. All the registered members own their own cars, and all they have seen other community members participating in different type social responsibility activities in the car forums, including environmental (45.1%), medical aid donation (18.7%), natural disaster relief donations (18.2%), school sponsorship (10.8%), and others (7.2%). Meanwhile, these members are not just bystanders; except a few (10.9%), most of them (89.1%) have been involved in community social responsibility activities. Specifically, the most common type of BCSR activity is environmental (26.9%), followed by medical donations (23%) and natural disasters (17.6%), followed by school sponsorship (12.1%) and other types of CSR activities (9.5%). In addition, there are two types of car brand forums, the majority of which are initiated by the members themselves (64.2%), and a small number of which are created by car companies or auto 4S stores (35.8%). The length of registered members of these automobile brand community members is mainly less than 1 year (5%), 1–2 years (35.4%) and 2–3 years (27.7%), and some are 3–4 years (15%) and more than 4 years (16.9%). Table 1 presents the demographic characteristics of the valid sample.

Table 1.

Descriptive statistics of the valid sample

Variable	Frequency	Percentage (%)	Variable	Frequency	Percentage (%)
Gender			Registered member or not
Male	246	45.6	Yes	539	100
Female	293	54.4	Own a car or not
Age (Years old)			Yes	539	100
≤ 20	3	0.6	Forum categories
21–30	302	56	X car company or 4S shop formation	193	35.8
31–40	213	39.6	Car club members	346	64.2
41–50	17	3.1	See members participate in public welfare events or not
> 50	4	0.7	No	0	0
Income (RMB)			Environmental	243	45.1
< 3000 Yuan	33	6.1	School sponsorship	58	10.8
3000–5999 Yuan	59	10.9	Medical aid donation	101	18.7
6000–8999 Yuan	135	25.0	Natural disaster relief donations	98	18.2
9000–11,999 Yuan	126	23.4	Others	39	7.2
12,000–14,999 Yuan	100	18.6	Participate in public welfare events or not
> 15,000 Yuan	86	16.0	No	59	10.9
Education			Environmental	145	26.9
High school or below	4	0.7	School sponsorship	65	12.1
Junior college	27	5.0	Medical aid donation	124	23
Undergraduate	447	82.9	Natural disaster relief donations	95	17.6
Master	52	9.6	Others	51	9.5
Doctor and above	9	1.7	Length of time to join a forum (Year)
Occupation			(0, 1]	27	5
Enterprise staff	71	13.2	(1, 2]	191	35.4
Full-time student	431	79.9	(2, 3]	149	27.7
Self-employed	21	3.9	(3, 4]	81	15
Staff of government organs and institutions	15	2.8	(4, 5]	44	8.2
Others	1	0.2	(5, 50]	47	8.7

N = 539

Measures

We investigated four constructs in the hypothesized model by using multi-item scales that have been employed in other studies. BCSR refers to brand community contributions such as environmental protection actions and donations; four items for measuring BCSR were adapted from CSR items (Currás-Pérez et al. 2009). Brand community collective self-esteem means that members are proud of being part of the community and comprises four dimensions: private collective self-esteem, public collective self-esteem, membership collective self-esteem, and importance to identity collective self-esteem (Luhtanen and Crocker 1992). The relationships among the four dimensions and overall collective self-esteem indicate that the dimensions are formative indices of collective self-esteem because when the dimensions are enhanced, collective self-esteem is strengthened. Brand community commitment is the degree to which a member desires to maintain a long-term relationship with the community and was measured using five items (Mathwick et al. 2008). Altruism is willingness to help others and was measured using three items (MacKenzie et al. 1993). The scales for measuring brand purchase and recommendation intentions were adapted from Chung and Kim (2020). Brand prestige refers to the market evaluation of a brand received after spreading (Bhat and Bowonder 2001), and the construct was measured using three items (Currás-Pérez et al. 2009). Community age (Compared with other car brand club forums, the development history of X car club forum is quite long) and size (Compared with other car brand club forums, the membership of X car club forum is very large) were each assessed using one item. Demographic variables such as gender, age, income, and education level were used as control variables. All latent variables were measured using a Likert scale ranging from 1 for strongly disagree to 7 for strongly agree.

Results

Tests of data reliability and validity

IBM SPSS Amos 26.0 was used in this study. The reliability test addressed internal consistency reliability and composite reliability. The results show a good fit to the data: (χ² = 605.30, df = 254, χ²/df = 2.383, CFI = 0.925, GFI = 0.914, IFI = 0.926, TLI = 0.912, RMSEA = 0.051). As shown in Table 2, all the items in the model were significant and the factor loading was greater than 0.5; all Cronbach’s α values for the constructs exceeded 0.7 except that for the BCSR(0.688), private collective self-esteem(0.686), public collective self-esteem(0.665), altruism (0.624), brand prestige (0.661), and brand recommendation intentions (0.649); the composite reliability of all constructs exceeded 0.8; thus, it confirmed that the items had adequate reliability (Hair et al. 2020). Validity testing included testing of discriminant and convergent validity. Table 2 reveals that all AVEs exceeded 0.5, which indicated that each measurement had favorable convergent validity (Hair et al. 2020). Fornell and Larcker (1981) suggested that each construct’s AVE should be compared to the squared inter-construct correlation (as a measure of shared variance) of that same construct and all other reflectively measured constructs in the structural model, and the shared variance for all model constructs should not be larger than their AVEs. Table 3 indicated that the coefficients of correlation between each construct did not exceed the square root of the AVE of each latent variable (Fornell and Larcker 1981). Thus, the measurement items had adequate validity.

Table 2.

Measurements and loadings

Measured items	Loading
Brand community social responsibility (Cronbach’s α = 0.688; CR = 0.811)
Brand Community X is aware of environmental issues	0.735
Brand Community X fulfills its social responsibilities	0.778
Brand Community X gives back to society	0.727
Brand Community X acts in a socially responsible way	0.634
Brand community collective self-esteem (Cronbach’s α = 0.802; CR = 0.871)
Membership collective self-esteem (Cronbach’s α = 0.766; CR = 0.851)
I am a worthy member of Brand Community X I belong to	0.793
I feel I don’t have much to offer to Brand Community X I belong to. (R)	0.713*
I am a cooperative participant in Brand Community X I belong to	0.721
I often feel I’m a useless member of Brand Community X. (R)	0.836*
Private collective self-esteem (Cronbach’s α = 0.686; CR = 0.809)
I often regret that I belong to Brand Community X I do. (R)	0.717*
In general, I’m glad to be a member of the Brand Community X I belong to	0.762
Overall, I often feel that Brand Community X of which I am a member are not worthwhile. (R)	0.737*
I feel good about the Brand Community X I belong to	0.652
Public collective self-esteem (Cronbach’s α = 0.665; CR = 0.800)
Overall, the Brand Community X that I am a member of is considered good by others	0.691
Most people consider Brand Community X, on the average, to be more ineffective than other social groups	0.699
In general, others respect Brand Community X that I am a member of	0.639
In general, others think that Brand Community X I am a member of is unworthy. (R)	0.794*
Importance to identity collective self-esteem (Cronbach’s α = 0.84; CR = 0.893)
Overall, Brand Community X memberships have very little to do with how I feel about myself. (R)	0.791*
Brand Community X I belong to is an important reflection of who I am	0.811
Brand Community X I belong to is unimportant to my sense of what kind of a person I am. (R)	0.846*
In general, belonging to Brand Community X is an important part of my self-image	0.840
Brand community commitment (Cronbach’s α = 0.791; CR = 0.857)
I would feel a loss if Brand Community X was no longer available	0.647
I really care about the fate of Brand Community X	0.738
I feel a great deal of loyalty to Brand Community X	0.764
The relationship I have with Brand Community X is one I intend to maintain indefinitely	0.778
The relationship I have with Brand Community X is important to me	0.760
Altruism (Cronbach’s α = 0.624; CR = 0.797)
Helps orient new agents even though it is not required	0.666
Is always ready to help or to lend a helping hand to those around him/her	0.743
Willingly gives of his/her time to help others	0.843
Brand prestige (Cronbach’s α = 0.661; CR = 0.815)
The people around me have a positive image of Brand Community X	0.773
In general, Brand Community X is a respected brand	0.792
Brand Community X is a brand with a good reputation	0.749
Brand purchase intention (Cronbach’s α = 0.806; CR = 0.885)
I intend to keep purchasing Brand X	0.862
My choice will be Brand X if I need to buy cosmetic products	0.841
I am willing to buy Brand X	0.842
Brand recommendation intention (Cronbach’s α = 0.649; CR = 0.811)
I would recommend Brand X to my friends	0.721
I am willing to promote Brand X in Brand Community X	0.794
I am willing to post and share the information about Brand X	0.784

Bootstrapping based on n = 5000

CR composite reliability, AVE average extracted variance. (R) means a reverse question

*The loadings of reverse questions are calculated as the response data of the reverse question minus 7

Table 3.

Convergent validity and discriminant validity

Variables	AVE	Discriminant validity
BCSR	0.519	0.72
IICSE	0.676	0.41	0.822
MCSE	0.589	0.518	0.565	0.767
PrCSE	0.516	0.498	0.394	0.515	0.718
PuCSE	0.501	0.563	0.462	0.493	0.655	0.708
BCC	0.546	0.572	0.555	0.522	0.505	0.508	0.739
BPI	0.72	0.399	0.307	0.267	0.359	0.334	0.462	0.848
BRI	0.589	0.517	0.382	0.418	0.438	0.455	0.605	0.573	0.767
BP	0.569	0.466	0.25	0.349	0.396	0.334	0.431	0.341	0.38	0.772
Altruism	0.596	0.455	0.41	0.518	0.498	0.563	0.404	0.431	0.34	0.377	0.754
CA	–	0.28	0.299	0.251	0.156	0.319	0.408	0.291	0.349	0.257	0.33	1
CS	–	0.372	0.273	0.3	0.265	0.339	0.399	0.382	0.416	0.325	0.411	0.562	1
MT	–	0.135	0.201	0.177	0.098	0.155	0.207	0.101	0.125	0.109	0.09	0.189	0.175	1
FC	–	− 0.002	− 0.013	− 0.044	0.042	− 0.018	0.009	− 0.044	− 0.019	− 0.046	0.045	− 0.052	− 0.053	− 0.002	1
WTAPWE	–	0.040	0.068	0.061	0.049	0.017	0.045	0.078	0.107	0.085	− 0.004	0.112	0.106	0.137	0.046	1

BCSR brand community social responsibility, BCC brand community commitment, BPI brand purchase intentions, BRI brand recommendation intentions, BP brand prestige, CA community age, CS community size, MT membership tenure, FC forum categories, WTAPWE whether to participate in public welfare activities

Bold numbers on the diagonal denote square roots of the AVE. The data in the lower-left-hand corner are coefficients of correlation between factors

Common methodological deviations

The data were reported by the respondents; thus, common method bias (CMB) among the variables. This paper describes the most common or recommended approaches that can be used to identify or control CMV impacts from the PLS-SEM study’s results. First, Bagozzi et al. (1991) described the method of assessing the impact of CMV through latent variables’ correlations; the common method bias will be evident when a substantially large correlation is found among principal constructs (r > 0.9). Except for the control variables, the correlation coefficients of the variables ranged from − 0.014 to 0.808 (< 0.9), and thus, the CMB was not severe (Bagozzi et al. 1991). Second, Chin et al. (2013) introduced a measured latent marker variable (MLMV) method to detect and correct CMV while utilizing partial least squares, which includes two approaches (i.e., construct level correction, CLC; item-level correction, ILC) can remove the impact of CMV on structural paths. The first approach called construct level correction (CLC) was used to determine whether CMB was present in this study. Specifically, this study has four constructs of the research model. Therefore, we draw four CMV control constructs involving 4 items of social desirability (unrelated measured items). The CMV control constructs were modeled to have an impact on each PLS-SEM model’s construct. Then, path coefficients are again estimated after introducing CMV control constructs on models’ constructs. It was observed that both original and CLC estimated result changes were very small and not significant, and thus, we can conclude that CMV was not any issue in this study. Likewise, original and CLC estimated t-values or R² values in the PLS-SEM model were non-significant changes. Therefore, after removing the impacts of common method bias by the CLC approach, the t-values of path coefficients were still significant (See Table 4). Moreover, these changes were not significant which also indicated that common method bias is not a serious issue in this study. Overall, the tests indicated that CMB was not detrimental to our results.

Table 4.

Comparison of R² values, path coefficients, and T-values by CLC approach and original PLS-SEM models

Constructs/relationships	R2-value estimation		Path coefficient estimation		T-value estimation
	CLC	Original	CLC	Original	CLC	Original
BCC	0.480	0.478
BCCSE	0.408	0.401
BPI	0.218	0.214
BRI	0.366	0.365
BCC → BPI			0.452	0.462	10.464***	11.313***
BCC → BRI			0.601	0.604	17.69***	18.854***
BCSR → BCC			0.235	0.237	4.259***	4.196***
BCSR → BCCSE			0.552	0.562	14.816***	15.185***
BCCSE → BCC			0.432	0.438	9.534***	9.062***
Altruism → BCCSE			0.052	0.055	2.502*	2.535*
Altruism → BCC			0.003	0.005	0.066ns	0.124ns

BCSR brand community social responsibility, BCCSE brand community collective self-esteem, BCC brand community commitment, BPI brand purchase intentions, BRI brand recommendation intentions

*p < 0.05, **p < 0.01, ***p < 0.001 (two-tailed). ^nsnot significant

Hypothesis tests

Main effects and controlling effects

After confirming the reliability, validity, and CMV of the measures, we tested H1, H2, H3, H6 and H7 with the conceptual model (Table 5). The overall model fit statistics are χ² = 430.055, df = 147, χ²/df = 2.987, CFI = 0.922, GFI = 0.918, IFI = 0.922, TLI = 0.909, RMSEA = 0.061, indicating an overall acceptable fit between data and the conceptual model. The results indicated that BCSR had a positive effect on brand community commitment (β = 0.421, t = 4.05, p < 0.001), supporting H1. BCSR also had a positive effect on brand community collective self-esteem (β = 0.808, t = 9.927, p < 0.001), and brand community collective self-esteem had a positive effect on brand community commitment (β = 0.449, t = 4.620, p < 0.001), supporting H2 and H3. BCC had a positive effect on brand purchase intentions (β = 0.675, t = 11.01, p < 0.001) and brand recommendation intentions (β = 0.598, t = 10.61, p < 0.001), supporting H6 and H7.

Table 5.

Results of the main effects and controlling effects

Model effects	Hypothesized paths	SPC	T-value	P value
Main effects	H1: BCSR → BCC	0.421	4.050**	0.000
	H2: BCSR → BCCSE	0.808	9.927***	0.000
	H3: BCCSE → BCC	0.449	4.620***	0.000
	H6: BCC → BPI	0.675	11.01***	0.000
	H7: BCC → BRI	0.598	10.61***	0.000
Controlling effects	Brand prestige → BCCSE	0.098	2.427*	0.016
	Brand prestige → BCC	0.158	3.990***	0.000
	Community age → BCCSE	0.125	2.997**	0.003
	Community age → BCC	0.143	3.449**	0.001
	Community size → BCCSE	0.061	1.403ns	0.161
	Community size → BCC	0.147	3.447**	0.001
	Membership tenure → BCCSE	0.140	3.986***	0.000
	Membership tenure → BCC	0.128	3.597***	0.000
	Forum categories → BCCSE	− 0.033	− 0.941ns	0.347
	Forum categories → BCC	0.008	0.230ns	0.818
	Whether to participate in public welfare activities → BCCSE	0.143	4.128***	0.000
	Whether to participate in public welfare activities → BCC	0.061	1.713ns	0.087

SPC standardized path coefficient, BCSR brand community social responsibility, BCCSE brand community collective self-esteem, BCC brand community commitment, BPI brand purchase intentions, BRI brand recommendation intentions

p < 0.05, **p < 0.01, ***p < 0.001 (two-tailed). ^nsnot significant

In this model, brand community collective self-esteem was influenced by four control variables: brand prestige, community size, membership tenure, and whether to attend public welfare events. Brand prestige was positively correlated with brand community collective self-esteem (β = 0.098, t = 2.427, p < 0.05), meaning that member evaluation of a brand significantly influenced brand community collective self-esteem. Community age and membership tenure also significantly influenced brand community collective self-esteem (β = 0.125, t = 2.997, p < 0.01; β = 0.140, t = 3.986, p < 0.001) because age refers to the degree of a community’s historical development and positively affects the pride of members, and the longer members stay in the community, the more pride they feel. Whether to attend public welfare events significantly influenced brand community collective self-esteem (β = 0.143, t = 4.128, p < 0.001) because participation in public welfare events enhances members’ sense of identity and belonging to the brand community. Brand community commitment was affected by four control variables: brand prestige, community size, community age, and membership tenure. Brand prestige significantly influenced community commitment (β = 0.158, t = 3.990, p < 0.001), meaning that brand prestige can actively promote brand community and enhance community commitment. Community age and membership tenure significantly influenced community commitment (β = 0.143, t = 3.449, p < 0.01; β = 0.128, t = 3.597, p < 0.001), demonstrating that a long community’s history boosts people were more likely to maintain a long-term relationship with a community and more members are significantly higher willingness to remain in the community. Community size also significantly influenced community commitment (β = 0.147, t = 3.447, p < 0.01), indicating that large communities can provide members with additional informational and social value as well as willingness to remain in the community.

Mediation, moderation, and moderated mediation effects

IBM SPSS Statistics 27.0 and SPSS PROCESS macro 3.3 software were used to test mediation, moderation, and the moderated mediation hypotheses (Model 8). To test these relationships, we adopted the bias-corrected percentile Bootstrap method and the theoretical model was tested by estimating the 95% confidence interval (CI) with 5000 sampled with repetition. If the 95% bootstrapped confidence interval of bias correction for the indirect effect contains 0, no mediating or moderating effect exists.

First, this paper examined the mediating role of BCCSE in the relationship between BCSR and BCC. As can be seen from the results (see Table 6), the direct influence of BCSR on BCCSE (β = 0.610, SE = 0.044, p < 0.001) as well as BCCSE on BCC was also significant (β = 0.538, SE = 0.055, p < 0.001); the confidence interval for the indirect effect of BCSR on BCC excluded zero (95% CI [0.200, 0.347]), and the direct effect was also significant (β = 0.282, SE = 0.048, p < 0.001). As expected, this meant that brand community collective self-esteem played a partial mediating role between BCSR and BCC. The indirect and direct effects had the same direction (a × b × c was positive). Thus, the mediation was complementary (Zhao et al. 2010) and H4 was validated. The BCCSE is a multidimensional construct. This study used the second-order latent structure of BCCSE to include all four dimensions in the analysis, and mediation tests should be conducted for each dimension. Thus, we explored the four dimensions of CSE in groups to show the differences in dimensions. The results (see Table 7) showed that all four dimensions of BCCSE had complementary mediating effects between BCSR and BCC, consistent with the mediating effect of viewing BCCSE as a holistic construct. However, according to previous scholarly research paradigms (e.g., Chen et al 2021), taking into account the themes presented in this study and the simplicity of the model, we chose to examine collective self-esteem as a whole to explore its mediating role between BCSR and BCC.

Table 6.

Results of moderated mediation analyses

Predictor variables	Outcome variables
	BCCSE(M)				BCC(Y)
	SPC	SE	P	95%CI	SPC	SE	P	95%CI
BCSR(X)	0.610	0.044	0.000	[0.525, 0.696]	0.282	0.048	0.000	[0.187, 0.377]
Altruism(W)	0.138	0.042	0.001	[0.553, 0.220]	0.178	0.040	0.000	[0.099, 0.257]
BCSR(X) × Altruism(W)	0.135	0.041	0.001	[0.545, 0.215]	− 0.014	0.039	0.715	[− 0.091, 0.067]
BCCSE(M)					0.538	0.055	0.000	[0.430, 0.062]
R2	0.359					0.473
F	99.73***					119.67***

SPC standardized path coefficient, BCSR brand community social responsibility, BCCSE brand community collective self-esteem, BCC brand community commitment

*p < 0.05, **p < 0.01, ***p < 0.001 (two-tailed)

Table 7.

The mediation tests of BCCSE each dimension

Path Relationships	SPC	SE	P	95% CI
BCSR(X) → MCSE (M) → BCC(Y)	0.083	0.026	0.002	[0.035, 0.137]
BCSR(X) → PrCSE (M) → BCC(Y)	0.112	0.037	0.003	[0.046, 0.187]
BCSR(X) → PuCSE (M) → BCC(Y)	0.120	0.038	0.002	[0.047, 0.188]
BCSR(X) → IICSE (M) → BCC(Y)	0.122	0.027	0.000	[0.075, 0.180]

BCSR brand community social responsibility, BCC brand community commitment, IICSE importance to identity collective self-esteem, MCSE membership collective self-esteem, PrCSE private collective self-esteem, PuCSE public collective self-esteem

Second, outcomes of the moderation tests (see Table 6) indicate that altruism has a positive relationship with BCCSE (β = 0.138, SE = 0.042, p < 0.01) and BCC (β = 0.178, SE = 0.040, p < 0.001). We introduced the interactive item of altruism and BCSR into the model, and the results showed that altruism had a significant moderating effect on the relationship between BCSR and BCCSE (β = 0.135, SE = 0.041, p < 0.01; 95% CI [0.545, 0.215], excluded zero) whereas the relationship between BCSR and BCC was not significant (β =  − 0.014, SE = 0.039, p > 0.05; 95% CI [− 0.091, 0.067], contained 0), which meant that H5a was supported, but H5b was rejected.

Third, we examined the difference in the indirect effect of BCSR on BCC through BCCSE between “high” and “low” values of altruism by using the PROCESS macro for SPSS (Hayes 2013). The results suggested (see Fig. 2, Table 8) that BCSR is positively related to BCC through BCCSE when altruism is “high” (β = 0.706, SE = 0.057; 95% CI = [0.593, 0.818], excluded zero), and the indirect relationship is also significant when altruism is “low” (β = 0. 515, SE = 0.047; 95% CI = [0.424, 0.607], excluded zero). In addition, the difference in indirect effect between “high” and “low” values of altruism is significant (β = 0.084, SE = 0.025; 95% CI = [0.037, 0.1134], excluded zero). Thus, the influence of BCSR on BCC is a mediated moderation effect.

Fig. 2.

Interactive effect of BCSR and altruism on BCCSE. Notes: BCSR, Brand Community Social Responsibility; BCCSE, Brand Community Collective Self-esteem

Table 8.

Conditional indirect effect results

Altruism	BCSR(X) → BCCSE(M)			BCSR(X) → BCCSE(M) → BCC(Y)
	SPC	SE	95% CI	SPC	SE	95% CI
Low (− 1 SD)	0.515	0.047	[0.424, 0.607]	0.227	0.034	[0.165, 0.297]
High (+ 1 SD)	0.706	0.057	[0.593, 0.818]	0.311	0.044	[0.229, 0.402]
Difference				0.084	0.025	[0.037, 0.134]

“High”/ “Low” = one SD above/below the mean. Unstandardized coefficients (and SEs) are reported. CIs are derived from 5000 bootstrapped resamples

SPC standardized path coefficient, BCSR brand community social responsibility, BCCSE brand community collective self-esteem, BCC brand community commitment

Discussion

Theoretical implications

BCSR is a new phenomenon in the field of brand community and is vital to the development of a brand community. This study discussed the mechanism through which BCSR influences brand community commitment and analyzed the mediating effect of brand community collective self-esteem as well as the moderating effect of altruism. The empirical results, obtained through PLS-SEM, suggest that (a) BCSR enhances brand community commitment, which promotes brand loyalty behavior (purchase and recommendation intentions); (b) brand community collective self-esteem exerts a partial mediating effect between them, and all four dimensions of the brand community collective self-esteem had complementary mediating effects between BCSR and brand community commitment; (c) altruism moderates the relationship between BCSR and brand community collective self-esteem rather than the relationship between BCSR and brand community commitment. Our analysis demonstrated that social responsibility initiatives give a community a positive reputation, which encourages members to perceive collective self-esteem, leading to community commitment. Furthermore, the degree of collective self-esteem differs among members. Altruistic members are likely to support social responsibility activities and to perceive brand community collective self-esteem. Thus, altruism positively moderates the relationship between BCSR and collective self-esteem. However, the hypothesis that altruism moderates the relationship between BCSR and brand community commitment was empirically rejected. A possible explanation is that, regardless of altruism, when a brand community (as a type of informal organization) assumes social responsibility, nearly all members highly appraise the community and subsequently commit to it. We also obtained no evidence supporting the connection between altruism and community commitment. Therefore, altruism does not play a moderating role in the relationship between BSCR and community commitment.

Although researchers have investigated CSR in a small business community (Worthington et al. 2006) and how CSR supports the community surrounding the company (Kleinrichert 2008), few have focused on the effect of CSR on a brand community, which is an informal organizational cluster of brand admirers (Chaudhry and Krishnan 2007). The current study enriches the literature on brand community and social responsibility as follows. First, it advances the BCSR concept to explore social responsibility from the brand community perspective, broadening the social responsibility research realm. A recent study indicated a paradox of CSR standards (De Colle et al. 2014). In contrast to CSR, BCSR is performed by an informal group comprising brand admirers. BCSR also differs from CSR in motivation, characteristics, and the effect of social responsibility initiatives. However, studies have neglected this difference. The BCSR phenomenon is becoming increasingly common and has a crucial effect on brand community commitment. Second, this study indicates that BCSR is a new approach to triggering brand community commitment. Prior studies have investigated the antecedents of brand community commitment (Kuo and Feng 2013; Wang et al. 2019). The present study demonstrated that social responsibility activities undertaken by brand community members affect the commitment of other members of the community. This finding enriches the research on the antecedents of brand community commitment. The mediating effect of collective self-esteem and the moderating effect of altruism elucidate the transition from BCSR to brand community commitment. Additionally, the results that brand community commitment affects brand loyalty behavior (purchase and recommendation intentions) are consistent with the previous findings that brand community relationships can translate into marketing effects such as purchase intention and word-of-mouth (Demiray and Burnaz 2019).

Managerial implications

Our research findings have several practical implications for those managing brand communities. First, although BCSR is an initiative that brand communities often launch themselves, related companies can also participate and offer support (e.g., money and materials). A brand community is an informal nonprofit organization that draws members from various locations; thus, organizing social activities such as environmental protection tasks and assisting impoverished students or those in disaster areas is difficult. The company behind the brand can increase the size and age of the brand community and retain members for a long time as well as provide labor, money, and resources to strongly support the BCSR and can participate in social activities such that the company combines its name with that of the brand community. Such activities enlarge the effects of the brand and the brand community in society, help the brand community fulfill its social responsibility, and ensure the additional commitment of members to the brand community, thereby stimulating and promoting members’ brand loyalty behavior, such as brand purchase and recommendation intentions, enhancing brand equity and brand competitiveness.

Second, companies should encourage altruistic behavior among members. Although altruistic impulses are not shared by all members, their altruism can be enhanced to some degree. For example, companies can encourage members who exhibit altruistic behavior inside or outside the community by promoting them as examples and giving them gifts, which will prompt other members to help others within or outside the community. Constructing a favorable altruistic atmosphere is crucial for community development.

Third, corporations should enhance collective self-esteem in brand communities. Our findings indicate that brand prestige influences the collective self-esteem of community members. Therefore, corporations should construct a positive brand image and use the media to spread reports about the good deeds of brand communities and use the media to spread reports about the good deeds of brand communities to gain a reputation that will make brand community members proud of their community, thereby enhancing community collective self-esteem. Furthermore, both community age and membership tenure affect members’ evaluations of the brand community. Therefore, enterprises should manage their brand communities with as much care as they do the businesses, put in place a series of measures to increase the cohesiveness and stickiness of their members, and improve their identification with their brand communities. In addition, the companies or brand communities can also actively encourage community members to participate in public welfare events, which can also promote the community collective self-esteem.

Limitations and further research

This study has some limitations. First, BCSR, CSR, and BSR involve organizational social responsibilities and share numerous similarities; however, a brand community is an informal organization, whereas corporations and brand-based nonprofit organizations are formal organizations. Whether our results can serve as a direct reference for CSR is uncertain and requires further research. The research questions that arise from our findings include the following: (a) Can a BCSR also have negative effects on a brand, e.g., if the BC strongly engages in a topic, but the brand does not? Or if community members exert more efforts and money into social causes instead of brand-related activities and purchases? Do mediate and moderating effects (e.g. initiative-related or brand-related) exist in the mechanism? Research on these topics will build bridges between brands and brand communities. (b) Why BCSR happens in the first place and how the phenomenon can be influenced by corporate actions and other uncontrollable factors? What is the upstream situation of BCSR? Is BCSR diffusible? How can managers attract members to BCSR activities? Does it matter whether the brand/company actively pushes the BCSR efforts or not (level of pro-activeness among the brand community)? Does it matter whether a relatively large group of brand followers are organized for BCSR or not (group effort vs. individual effort)? (c) What types of members are likely to participate in social responsibility activities? What are the differences in backgrounds between participants and nonparticipants? Do those who participate in the CSR initiatives show a higher level of commitment compared to those who stay aside to observe these actions in the brand community (participation vs. observation)? Do those who contribute time show a similar level of brand commitment compared to those who donate money (time vs. money)?

Biographies

Zhimin Zhou

is a Professor at the College of Management at Shenzhen University, China. He received his PhD in Marketing from Sun Yat-sen University, China, in 2003. His research interests focus on brand management. He has published more than 80 papers in some prestigious journals, including Journal of the Academy of Marketing Science, Journal of Product & Brand Management, Journal of Computer-Mediated Communication, Journal of Business Research, Cyberpsychology, Behavior, and Social Networking, Internet Research, Journal of Destination Marketing & Management, and some leading Chinese journals.

Yucheng Wang

is a research assistant at the School of Business at Hong Kong Baptist University, Hong Kong. He received his master in Marketing from Shenzhen University, China in 2022. His research interests focus on consumer psychology and behavioral decisions. He has published some English papers in Journal of Product & Brand Management and Current Issues in Tourism as well as three papers in some Chinese journals.

Yaqin Zheng

is an assistant professor at the School of Business at University of Shanghai for Science and Technology, China. She received her PhD in Business Administration from Nanjing University, China, in 2016. Her research interests focus on sales research and brand management. She has published some English papers in Industrial Marketing Management, Journal of Business and Industrial Marketing, Global Strategy Journal and some leading Chinese journals.

Shixiong Liu

is a professor and Head of Department of Marketing, Shenzhen University, China. He received his PhD in Marketing from Sun Yat-sen University, China, in 2004. He is interested in researching consumer behavior, advertising and brand management. He published some books as well as papers in Journal of Business Research, Journal of Retailing and Consumer Services, Frontiers in Psychology, Journal of Consumer Marketing and some leading Chinese journals

Funding

Funding was provided by National Natural Science Foundation of China (Grant Nos. 72172093, 71832015), Stable Support Project of Shenzhen (Grant No. 20200813110000001).

Declarations

Conflict of interest

The authors declare that they have no conflict of interest.