The Relationship between CSR Communication on Social Media, Purchase Intention, and E-WOM in the Banking Sector of an Emerging Economy

Communication the Abstract: Modern businesses in the present digital age are facing different challenges for survival and growth due to the stiff competitive environment that prevails almost in every sector. Technology, especially digital technology, has changed the way of doing business around the globe. Marketers need to develop new strategies by incorporating the element of digital technology to overrun their competitors, as conventional competitive strategies will not produce extraordinaryresults. The rise of social media, in this regard, is a game-changer in the context of marketing as it provides a strategic touchpoint to the marketers to involve the customers with a brand. Prior research has largely ignored the relationship between corporate social responsibility (CSR) practices of a brand and positive customer experiences. Hence the present study aims to test the relationship of customer-related CSR activities on social media on behavioral consequences of customers, such as their purchase likelihood and electronic word of mouth (E-WOM). The data were collected from the banking customers of a developing economy and were analyzed through structural equation modeling in AMOS software. The results revealed that customer-related CSR activitieson social mediainﬂuence customer behavioral outcomes like E-WOM and purchase intentions. The result also revealed that brand attitude mediates the relationship between customer-related CSR, E-WOM, and purchase intentions. The ﬁndings of the present study can help policymakers to understand the importance of CSRpractices from the perspective of marketing, which is largely ignored by contemporary CSR researchers.


Introduction
Social media with its interactive feature and open-source features has become a prominent place for customers to engage with organizational relationship management including corporate social responsibility (henceforth referred to as CSR) [1]. CSR has been recognized as an important part of the branding that benefits organizations through influencing customer attitude towards a specific brand [2]. Carroll [3] defines CSR as "responsibilities and obligations of the businesses to improve the well-being of the community through the use of business resources such as finance, personnel, and equipment". Contemporary researchers have long established that CSR can improve customer satisfaction [4], organizational performance [5], reputation [6], and product quality [7]. However, what has been missed in recent literature is the relationship of CSR activities with customer behavioral outcomes. Itis quite recent that modern researchersin the literaturehave started to recognize the importance of CSR to achieve marketing-related objectives [8,9]. This changing approach towards CSR is receiving the attention of contemporary researchers. Therefore, CSR has become a popular strategy for modern businesses around the globe for reaping multiple business objectives [10]. Further, the rise of social media, in the current digital age, has brought a paradigm shift in the field of marketing [11]. Unlike conventional media, social media is a double-edged sword because on one hand, it conveys marketing communication to the potential audience, and on the other hand, its interactive atmosphere makes it possible to engage customers with an organization [12].
Although many companies are trying to use social media for advertising purpose, an increasing number of brands have begun to explore using social media as a platform to engage customers for their CSRefforts [13]. Customer-related CSR activities on social media are defined as the degree to which social media users communicate with others aboutthecompany's CSR initiatives [14]. In other words, it refers to the involvement of customers in communicating company's CSR initiatives (for example, liking or sharing). The extant literature has investigated the impact of CSR in shaping customer behavior [4,15,16]. However, little is known about the role of social media in developing positive customer behavior through CSR activities.
According to a recent report, the social media users around the world reached 4.54 billion with an increase of 7% per year, that shows that the rate of acceptability and importance of social media is unmatched as compared to its counterpart, the conventional media [17]. However, contemporary researchers have failed to address how customers' CSR-related initiatives on social media can shape the attitudinal and behavioral preferences for a specific brand. Itis worth mentioning here that positive brand attitude of customers is a key concern for marketers because it plays a key role in shaping customer purchasing intentions or buying preferences [18]. A growing body of modern researchers have started to examine customer journey with organization as a foundation to explain customer experience, for instance the studies of Kuehnl, et al. [19] and Rodríguez-Sánchez, et al. [20] are recent contributions to the contemporary literature in this connection. Further, marketers need to realize the key touch points to engage the customer in order to induce their purchase intentions and other behavioral outcomes including positive word of mouth and advocacy behavior [21]. However, limited research studies have focused on key touch points that can make the customers' experience with organization meaningful, especially the importance of CSR practices and communicating it with customers to shape their behavior has been largely ignored in extant literature. Likewise, the emergence of social media, have been recognized asagame changer for engaging modern customers in a meaningful way [22,23]. There is growing concern among contemporary researchers to understand how recent businesses can leverage social media to highlight their sustainablity practices and enhancing meaningful customer experience [24,25].
With this background, the present study aims to bridge these gaps by examining the roleofcustomers' CSR-related activities, using a social media platform, to build positive perceptions on the part of customers. The positive perceptions, created by customer-related CSR activities on social media, lay down the foundation to shape positive attitude of customer for a brand. This study referred to itas brand attitude and defines it in line with the definition of Percy and Rossiter [26] who argued that "brand attitude as the evaluation of a brand, on the part of buyers, in line with the expected capacity of a brand to serve a particular buying motive". Customer's brand attitude is a key touch point to shape customer behavior in a positive manner. For example, prior studies have wellestablished that customers' brand attitude positively influence their purchase intentions [27,28] and positive word of mouth (henceforth referred to as WOM) [29]. Unlike previous studies, the study at hand aims to investigate the relationship of customer-related CSR activities, purchase intentions, and electronic word of mouth (henceforth referred to as E-WOM) in the context of social media, whereas previous studies examined such relationships using conventional media [30,31]. Hence, the present study argues that customer-related CSR activities on social media positively influence their E-WOM and purchase intentions, whereas brand attitude mediates between customer-related CSR, E-WOM, and purchase intentions.
The proposed model of the study is tested in the banking sector of Pakistan. The banking sector was considered as a target sector due to three specific reasons. First, the banking sector is labeled as a homogenized sector in which most of the banking institutions offer the same kinds of products or services to their customers, hence customer retention is a big question for any bank due to this homogenized character of this industry. Therefore, the proposed model of the present study will give hope to the policymakers, that through a well-planned CSR program for marketing related purpose, a particular bank can hold its customer portfolio as compared to the competing banks which do not establish CSR programs to retain their customers. Second, the selection of the banking sector of Pakistan also seems logical, because it is a sector in which an established CSR structure is visible, whereas in other business sectors of the country, CSR activities are less visible as compared toPakistan, and CSR is still at its infancy stage. Third, the banking sector in Pakistan is highly competitive, which poses a challenge for every bank about retaining customers in an industry that is already volatile. Hence the integration of CSR with marketing-related objectives may create a sort of competitive advantage.
The study at hand contributes to contemporary literature in many ways; for instance, it adds to the recent literature of CSR by looking at CSR with marketing perspective, whereas, prior studies mostly explored CSR strategies in the context of organizational performance [5,6]. In the same vein, the study will enrich the contemporary CSR literature in the context of positive customer behavior, as the present study argues that customerrelated CSR initiatives onsocial media promote positive perceptions among customers, which ultimately shape their behavior in positive terms by building positive brand attitude, WOM, and purchase intentions. Lastly, the present study adds to the scant literature of CSR in the domain of social media as the majority of previous studies have largely focused on conventional media in this regard. The remainder of this article is composed as follows:The coming section describes the literature review section in which hypotheses are derived and the research model is proposed. In the next section, the authors have explained methodology, which includes sample, data collection, and measurements; after this section, there istheresults sections in which the authors have presented the results of data analysis, and finally comes the discussion section of the present research study, which includes implications and limitations of present research study.

Literature Review and Theoritical Support
This study uses two theories such as the theory of norm reciprocity [32] and attribution theory [33]. These two theories have been extensively used by different researchers in extant literature related to the behavior of individuals [34][35][36][37][38]. These two theories are relevant in the context of present study because the process of attribution helps customers in evaluating a brand favourably or unfavorably. Therefore, when customers feel that a particular organization is concerned with socital issues and want to solve these issues through different CSR activities, their attribution process helps them in evaluating a brand positively. Hence, they are expected to show positive attitude towards brand. In a similarmanner, in terms of norm reciprocity, when customers observe an organization is participating in CSR activities, they acknowledge such positive moves of an organization and want to reciprocate it positively. Hence, they support such socially responsible brands positively and use positive WOM for such organization. All such actions also help them to induce their purchase likelihood for a socially responsible organization.
Customer-related CSR activities on social media are exercised between the organization and its key stakeholders including customers. Organizations use different online media platforms to establish dialogue with customers thatinclude organizational webpages, Facebook pages, contents on Twitter, etc. [22]. When an organization shares CSR-related communications with its stakeholder through social media, customers consider such an organization as a transparent entity that conducts its business with honesty and in a fair manner [39]. This belief of transparency and fairness develops positive perceptions among customers for the organization and in return they show positive brand attitude [40]. Customer-related CSR communication in thepresent digital age is an effective strategy from the perspective of marketing that binds customers with an organization emotionally and rationally [41]. Kollat and Farache [42] theorized that organizations can gain customer trust through customer journey on social media and sharing with them their CSR communications. Social media provides an interactive forum to the customers on which they can express their views about the organization and can share, like, or comment on different posts with others [43]. The interactive capability of social media develops a sense of association among customers for the brand, which leads them to develop positive attitude [44]. When customers observe that an organization is sharing its CSR-activities on social media, they feel positive about the organization as they convince themselves that the organization not only cares for its customers but also responds to society positively [45]. Hence, it is logical to propose:

Hypothesis 1 (H1):
Customer-related CSR activities on social media positively relate to brand attitude.
Customersmake purchasing decisions based on their organizational wellbeing and product analysis. If customers have a positive attitude about CSR actions of an organization, their appreciation of the organization is expected to be improved [4]. However, evidence shows that customers often do not understand CSRprograms due to their lowlevel understanding about CSR initiatives [46]. It is quite difficult to achieve the CSR effect with the intention of buying if they are not always taken into account. Customers' understanding of CSR actions will depend, at least in part, on the strength of the CSR information, communicated by the organization to the customers.
The rise of social media is a game changer in this perspective because it provides a platform to the organizations through which they not only communicate their CSR strategies with different stakeholders like customers but also highlight the importance of their CSR initiative for community and environment [47]. Further, customer-related CSR initiatives of an organization on social media are also important due to their interactive feature because customers and company interactively communicate with each other. This interactive atmosphere of social media plays a key role in building customer confidence on a specific organization, which in turn is translated into increased buying intentions of customers with the organization [48].
Prior studies of customer response and CSR have focused on behavioral response and perceived reaction, where the former is associated with external response and latter is attributed to internal response [49]. Perceived reaction is an internal response of customers and itdeals withhow CSR initiatives affect customer perceptions, and it also defines the evaluation of an organization at the part of customers in term of sustainable practices. Preliminary literature pertinent to perceptional reaction emerged in 1997 in a study conducted by Brown and Dacin [50], who noted that understanding from the customer perspective about an organization will affect evaluations of the quality of products and services of an organization. Behavioral responses are external responses of customers, which typically include the impact of CSR actions on customer behavior, such as buying behavior, loyalty, purchase intentions, and willingness to pay a price premium [4]. Customers are more likely to accept services and products from a trusted company, which is proactively involved in CSR activities [51]. In a nutshell, the present study argues thatwhen an organization engages its customers on social media by communicating CSR-activities, it is expected that customers will reciprocate such socially responsible behavior in a positive manner that ultimately influences their buying intentions. Hence, it is proposed:

Hypothesis 2 (H2):
Customer-related CSR activities on social media positively relate to purchase intentions.
The organizations participate in CSR operations in order to develop a positive attitude of the customers towards the organization and to build its positive brand image [52]. Incorporating a CSR program increases organization's visibility, and encourages more communication with customers. Therefore, it is expected that an organization's engagement into CSR activities is likely to induce positive WOM on the part of customers [53]. Customers are concerned about the quality of service provided by the enterprise, in terms of social responsibilities. CSR engagement of an organization builds positive image of the organization [54]. Through the halo effect, the positive image of an organization created by CSR activities leads customers to develop a better perception about the organization. This well-rounded perception of the organization attract customers for the organization's products and services [55]. As a result, customers often speak positively about the company that is engaged in CSR-initiatives, as well as associating themselves positively with the products and services of the organization. When customers observe an organization practicing CSR programs actively, they buildapositive attitude towards that organization and communicate positively, hence they spread positive WOM for the organization [56]. Positive WOM refers to a good communication about a company that the customer is willing to share with others [57]. When customers spread positive communication about an organization or brand thorough social media or internet, then it is regarded as E-WOM [58].
The role of positive WOM is well recognized among contemporary researchers for gaining new customers [57,59,60]. Further, it has been also recognized by recent researchers that positive WOM influences customer's positive brand preference [61], perception [29], buying intentions [62], and loyalty [63]. Therefore, WOM is very important for organizations that effectively use WOM to increase their sales or successful promotion. With the advent of the Internet, online media has facilitated WOM online communication. In addition, organizations participate in CSR for a number of reasons, includingpositive WOM [64]. As a result of CSR involvement, customers will be more willing to talk aboutorganization's CSR activities with their friends, family and partners. Therefore, organizations involved in CSRactivities build their image of being a socially responsible organization [65]. Modern organizations ofthepresent era use social media to communicate their CSR motives with customers and other stakeholder. Whencustomers on social media see such activities of an organization, they feel pride for that organization and also communicate about the organization positively with their social media peers [66]. Hence, it is proposed:

Hypothesis 3 (H3):
Customer-related CSR activities on social media positively relate to E-WOM.
Research has long established that customer-related CSR initiatives of an organization influence customer's brand attitude positively [40,67], because when customers realize that a particular organization is a socially responsible entity, they associate themselves with that organization on positive terms, whichinfluences their brand attitude for that organization. In line with the theory of reasoned action by Ajzen and Fishbein [68], the present study theorized that brand attitude positively affect E-WOM and purchase intentions of customer. Unlike traditional media, the emergence of social media has provided customers an interactive and flexible media where they can share their views about an organization and can spread viral messages [69]. The theory of reasoned action contends that customers' brand attitude is related positively with purchase intention, in the same vein, it can be expected that CSR activities of an organization are expected to influence customers' attitude towards brand, which in turn induce buying intentions of customers for a brand or organization [64].
There have been studies intherecent literature in which traditional theory of reasoned action is extended by the researchers on grounds such as transparency, trust, and social responsibility as determinants of behavioral intentions; for instance, Hussain, et al. [70] mentioned in their study thatthetheory of reasoned action is helpful in shaping behavioral intentions of customers including buying intention for a particular organization. Likewise, in a further study, Kang and Hustvedt [71] theorized that customers' belief that a particular organization is socially responsible and transparent directly influenced their attitude towards brand and indirectly influenced their purchase intention and E-WOM. Further, Spears and Singh [72] found that customer attitude towards brand is positively related to customer purchase intention. To conclude, this study contends that in the present digital age, the majority of customers use social media to be engaged with the organization's CSR activities. Such engagement is expected to shape their attitude towards the brand positively, which ultimately influences their likelihood to spread positive viral messages about an organization, hence customers' E-WOM is positively influenced by brand attitude and CSR activities. In a similarmanner, the study also argues that customer-related CSR activities on social media influence their purchase intentions via positive brand attitude. Therefore, it is logical to propose the following set of hypotheses.

Hypothesis 4 (H4):
Brand attitude mediates between customer-related CSR activities on social media and purchase intention.

Hypothesis 5 (H5): Brand attitude mediates between customer-related CSR activities on social media and E-WOM.
Furthermore, the authors have reported a summary from recent literature to furtherhighlight the novelty of the present study in tabular form (Table 1). The authors have comprehensively summarized the relevant literature in Table 1. As per the findings of Table 1, it is obvious that there are several researchers who carried out different CSR studies in different domains like the relationship of CSR with purchase intentions, brand attitude, and E-WOM. However, no study in Table 1 comprehensively observed how CSR communications of an organization on social media can be helpful to shape customers' brand attitude, E-WOM, and purchase intention. In this regard, the current study is a pioneer effort to test such effects under a single research model (Figure 1). Thus the novelty aspect of this study is clearly highlighted in the light of the above discussion.

Methods
The proposed model of the present study is aimed to be tested in the banking sector of Pakistan. To represent the banking sector, the authors selected the banks thatwere actively involved in CSR activities and they have also their CSR presence on social media. The initial assessment of authors resulted in the selection of five banks namely Habib bank limited (HBL), National Bank of Pakistan (NBP), United Bank Limited (UBL), Meezan Bank Limited, and MCB Bank. These selected banks actively use social media to engage different stakeholders, despite that they have also a presence in different countries other than Pakistan. Hence, in the perspective of the present study, the selection of these banks seems logical. The data were collected from the individuals who had a bank account in one of these banks in the city of Lahore, which is one of the largest cities in Pakistan. Further, there are multiple branches of these banks in Lahore city at different locations.
The respondents of present study were approached by the authors when they were leaving a particular branch of the bank or they were present outside of the ATM facility. The collection of data in this way is in line with the recommendation of Ahmad, et al. [79] and Sun, et al. [80]. Before actuating the data collection phase, the authors received the informed consent of the respondents to voluntarily participate in the survey. Further, the respondents were told that the secrecy of their data will be highly maintained and will

Methods
The proposed model of the present study is aimed to be tested in the banking sector of Pakistan. To represent the banking sector, the authors selected the banks thatwere actively involved in CSR activities and they have also their CSR presence on social media. The initial assessment of authors resulted in the selection of five banks namely Habib bank limited (HBL), National Bank of Pakistan (NBP), United Bank Limited (UBL), Meezan Bank Limited, and MCB Bank. These selected banks actively use social media to engage different stakeholders, despite that they have also a presence in different countries other than Pakistan. Hence, in the perspective of the present study, the selection of these banks seems logical. The data were collected from the individuals who had a bank account in one of these banks in the city of Lahore, which is one of the largest cities in Pakistan. Further, there are multiple branches of these banks in Lahore city at different locations.
The respondents of present study were approached by the authors when they were leaving a particular branch of the bank or they were present outside of the ATM facility. The collection of data in this way is in line with the recommendation of Ahmad, et al. [79] and Sun, et al. [80]. Before actuating the data collection phase, the authors received the informed consent of the respondents to voluntarily participate in the survey. Further, the respondents were told that the secrecy of their data will be highly maintained and will not be shared with anyone. The authors also cleared the respondents about the importance ofthepresent research and requested them to give their genuine responses to the survey items. The authors distributed 600 questionnaires among different respondents and finally received 372 questionnaires, which were used in the final set of data analysis. Hence, the response rate of the present study remained around 62%.
In order to reduce the issue of social desirability bias, the authors took different steps, for instance, the survey items were carefully cross-checked by the experts in the field, in order to avoid any ambiguity and suitability of the instrument to serve the objectives of the present study. Likewise, the survey items were randomly scattered into the questionnaire in order to break any intended sequence that may be developed on the part of respondents. In the same vein, the authors remained present during the phase of data collection in order to address any potential difficulty in filling the questionnaires on the part of the respondent. Next, the authors also addressed the potential issue of common method bias (CMB), as the data of the present study were collected from the same respondent at a given interval of time. Hence, although there is every likelihood of the presence ofCMB, the authors addressed this issue in light of the guidance ofHarman [81]; theauthors in this regard loaded all the study items onto a single factor in order to determineif there is any single dominant factor. The results of the single-factor analysis revealed that there is no such dominant factor as the most variance explained by a single factor was 32.6%, which indicated that CMB is not a potential issue in the data of the present study. The demographic profile of the respondents is presented in Table 2.

Measures
The present study used already established scales for all the variables of the study, for example the items of customer-related CSR activities on social media were taken from the study of Gil de Zúñiga, et al. [82]. The scale consisted offive items (a sample item: "I Post personal experiences related to my bank's CSR activities"). The scale of brand attitude was adapted from Muehling [83], which consisted of three items (a sample item: "My attitude toward my bank is positive"). Moreover, the items of E-WOM were adapted from the study of Kang and Hustvedt [71], and this scale also consisted of three items (a sample item: "I would recommend this bankto my friends on social media"). Finally three items of purchase intentions (a sample item: "I am likely to consider this bank next time I think about buying banking services") were taken from the study of Kang and Hustvedt [71]. All the items were rated on a five point likert scale.

Results
The authors started data analysis phase with factor analysis in SPSS software in order to determinewhether items of the survey werewell loaded on to their respective factors. The results are shown in Table 3, and revealed that there is no issue in factor loadings as all the items loaded on to their respective factors. The rule of thumb here is [84] if an item loads to its respective factor with a value of 0.5 or more it means the item is well loaded onto the factor. Item loading values closer to 1 are considered more accurate. Further, the values of standard deviation (SD) are also reported in Table 3, and all the values are less than 1, which means that the data arecloser to themean with less variability, which is a good indication to take the data analysis to afurther level. The results of correlation (see Table 3) analysis confirmed that the variables of the present study are positively correlated with each other; moreover, these values are in moderate ranges, which means that there is less likelihood of multicollinearity issue amongst variables. Next, the authors tested the results of confirmatory factor analysis in AMOS software in order to confirm whetherthe data fit the theoretical model of the present study (see Table 3 and Figure 2). The results revealed that the database model fits to the theoretical model nicely because the results of model fit indices were appropriate (χ 2 /df = 3.27, RMSEA = 0.511, NFI = 0.943, CFI = 0.949, GFI = 0.947). The authors also evaluated convergent validity, discriminant validity, and reliability of the instrument. To assess convergent validity, the authors calculated average variance extracted (AVE) for each variable and revealed that all values were above the cut-off level of 0.5. Likewise, the discriminant validity was assessed through the square root values of AVE for each variable, which was then compared to the relative correlation values. The results confirmed that all variables are distinct from each other. The reliability results were assessed by observing the values of Cronbach alpha (α) and composite reliability (CR), with both tests producing sufficient results (α > 0.7, CR > 0.7).Lastly the authors also presented the results of data normality, and in this regard, the authors used the guidelines ofBrown and Dacin [50] who recommended that the data arenormally distributed if the Skewnessvalue ranges between ±3, and Kurtosis between ±10.

Hypotheses Testing
The author tested the results of the hypothesis through the structural equation modeling technique (SEM) in AMOS software, which is a covariance-based analysis technique. For hypotheses testing, the authors ran the structural model in AMOS in two steps; in the first steps, the structural model was executed to test the direct effects without any mediator. The results revealed that hypotheses 1, 2, and 3 are significant and true as indicated by their respective beta values and p-values (H1; β = 0.211 **, LLCI = 0.177, ULCI = 0.419, p < 0.05-H2; β =0.231 **, LLCI = 0.287, ULCI = 0.721, p < 0.05-H3; β = 0.173 **, LLCI = 0.157, ULCI = 0.236, p < 0.05). Next, the authors usedthebootstrapping option in AMOS to test the results of mediation effects for hypotheses 4 and 5 (see Table 4). The results revealed that brand attitude is a partial mediator between CSR and PI, and between CSR and E-WOM (because the beta value is reduced from 0.231 ** to 0.033 ** for H4, and from 0.173 ** to 0.118 ** for H5). These results are also reported in the structural model (see Figure 3).

Hypotheses Testing
The author tested the results of the hypothesis through the structural equation modeling technique (SEM) in AMOS software, which is a covariance-based analysis technique. For hypotheses testing, the authors ran the structural model in AMOS in two steps; in the first steps, the structural model was executed to test the direct effects without any mediator. The results revealed that hypotheses 1, 2, and 3 are significant and true as indicated by their respective beta values and p-values (H1; β = 0.211 **, LLCI = 0.177, ULCI = 0.419, p < 0.05-H2; β =0.231 **, LLCI = 0.287, ULCI = 0.721, p < 0.05-H3; β = 0.173 **, LLCI = 0.157, ULCI = 0.236, p < 0.05). Next, the authors usedthebootstrapping option in AMOS to test the results of mediation effects for hypotheses 4 and 5 (see Table 4). The results revealed that brand attitude is a partial mediator between CSR and PI, and between CSR and E-WOM (because the beta value is reduced from 0.231 ** to 0.033 ** for H4, and from 0.173 ** to 0.118 ** for H5). These results are also reported in the structural model (see Figure 3).      CSR, customer-related CSR; BA, brand attitude; E-WOM, electronic word of mouth; PI, purchase intention; S.E = standard error, LLCI = lower limit confidence interval, ULCI = upper limit confidence interval. * model fit indices for direct effect model; ** beta value significant. *** model fit indices for mediated (three factor) model.

Discussion and Implications
The aim of the present study was to test the effect of customer-related CSR activities on social media on E-WOM and purchase intentions of the customers with the mediating effect of brand attitude. The statistical results of the study revealed that customer-related CSR activities are directly and indirectly, through brand attitude, related to E-WOM and purchase intentions of customers. The rise of social media in the present digital age is a game-changer for modern businesses to grab the attention of different stakeholders including customers. Unlike traditional media, which mainly lies in the one-way communication domain, social media is an interactive medium that provides greater flexibility and coverage to its users. It is very important for contemporary businesses to recognize the importance and power of social media especially to attain marketing-related objectives, by engaging customers on social media. Almost half of the world's population uses the internet in recent times [85] and customers on a daily basis receive a lot of information; in other words, customers have to absorb information overload on the internet that gives birth to the issue of virtual fatigue for customers. Hence, modern customers want to experience meaningful interaction with organizations and brands. Customer-related CSR activities on social media provide them a platform to interact meaningfully in the context of CSR practices. So, the findings of the present study confirmed that customer CSR engagement on social media positively relates to E-WOM and purchase intentions. This study is not the first one to present these arguments because the arguments also receive support from contemporary researchers including Badenes-Rocha, Ruiz-Mafé, and Bigné [47]; Bialkova and Te Paske [58]; and Chu and Chen [64].
The present study has some important implications for theory and practice. First, the study at hand enriches the prior literature in the domain of CSR with marketingrelated intervention such as it uses a different lens to look at CSR, as it tries to describe the sustainable practices of an organization from the perspective of customers, which is largely ignored in prior literature. Second, the present study also adds in the existing literature of sustainable practices from the perspective of an emerging economy whereas most of the prior literature focused on the issue at hand, in the context of developed countries. Lastly, another important theoretical implication of the present study is the inclusion of brand attitude in the proposed model as a mediator. The study enriches the contemporary literature by drawing upon the importance of CSR activities on social media as an agent of change in customer attitude that ultimately affects E-WOM and purchase intentions of the customers.
There are some implications of the present study for practice that cannot be ignored, for instance, this is the time for policymakers to change the conventional thinking about CSR, which assumes that CSR is mostly related to philanthropic activities, as the recent calls from contemporary literature [16,79] have confirmed that looking at CSR with a marketing lens can produce dual benefits for organizations including economic benefits and social benefits. Modern marketers need to realize new directions to meaningfully engaging customers in order to drive a better customer experience. The use of social media in this context is a dawn of a new horizon that engages different stakeholders like customers in CSR activities. When customers observe the sustainable practices of an organization, they associate themselves with the organization on positive terms, hence they spread positive E-WOM about that organization by sharing and liking CSR-related content of an organization. Hence, all these activities create a positive atmosphere for an organization created by the customers. The findings of the present study bring it to the notice of policymakers that they need to realize the importance of social media to purposefully engage the customers by communicating with them the different CSR activities.
The practical implications of the present study for the banking sector are of utmost importance due to two specific reasons. First, the banking sector is largely considered a homogenized sector in which it is assumed that all players are serving the customers in a standardized way without any observable differentiation. Hence, holding or retaining the customers remains, and will remain, a challenge for this sector due to its homogenized nature. Sustainable practices in this regard may provide an opportunity for a bank to differentiate itself from the rest of the crowd on the grounds of social responsibility because the modern customers are concerned with sustainable brands. Second, the findings of the present study are important for the banking sector of Pakistan to change their approach towards sustainable practices, because most of the banking sector in Pakistan invests in CSR activities with aphilanthropic approach, andalthough there is nothing wrong with this philanthropic approach, if economic benefit could also be achieved through well-planned CSR initiative, then it is not wise to ignore this perspective of sustainability for businesses including banking sector of Pakistan.

Limitations and Direction for Future
The present study is not without limitations, but the authors are convinced that these limitations also open new ways for upcoming researchers in the same field. The first limitation of the study lies with the proposed research model in which the authors only considered the positive aspect of E-WOM andthe negative aspect of E-WOM is ignored, which may be very important because it is an indication of customer dissatisfaction and disapproval. Hence, just looking at the positive side and ignoring the negative side of a construct may be risky, so upcoming researchers are required to incorporate this limitation. Second, this study addressed the "how" part of the research by confirming how customerrelated CSR activities on social media can inducebehavioral outcomes, but why customers consider social media to be engaged in CSR-related communication, and what motivates them to participate in such activities on social media, is still under-explored. So the future studies need to address this limitation of the present study. Lastly, the cross-sectional nature of data is also a potential limitation of the present study as it limits the strength of causality of the relationship among study variables, hence future researchers are required to use longitudinal data to address this limitation.
Author Contributions: All of the authors contributed to conceptualization, formal analysis, investigation, methodology and writing and editing the original draft. All authors have read and agreed to the published version of the manuscript.
Funding: This research has received funding from the National Research Project Reference PID2019-108503RB-I00.
Informed Consent Statement: Informed consent was obtained from the respondents of the survey.
Data Availability Statement: The data will be made available on request from the corresponding author.

Conflicts of Interest:
The authors declare no conflict of interest.