Bridging the gap between work- and nonwork-related knowledge contributions on enterprise social media: The role of the employee–employer relationship
Nabila Boukef, Mohamed Hédi Charki, and Mustapha Cheikh-Ammar contributed equally to this work.
Abstract
Knowledge is an invaluable resource and a key to organisational success. To leverage this resource adequately, organisations must encourage their employees to share what they know with their peers. Enterprise social media (ESM) has emerged as an ideal venue for achieving this goal, and numerous studies have examined the drivers of work-related knowledge contributions on these platforms. The present study contributes to this body of research by examining a prevalent yet underexplored form of knowledge sharing that often occurs on ESM: nonwork-related knowledge contributions. We argue that contrary to a commonly held belief, this presumably hedonic employee behaviour can benefit organisations through its spillover effect on the work domain. In other words, we argue that nonwork-related knowledge contributions on ESM can foster work-related ones. Building on social exchange theory and on the associative–propositional evaluation model in social psychology, we also show that the employee–employer (EE) relationship—conceptualised in terms of perceived organisational support and perceived employee psychological safety—moderates the relationship between the two forms of knowledge contributions. The analysis of field data collected from 269 employees of a French e-commerce company confirmed that nonwork-related knowledge contributions are positively associated with work-related ones and that this positive association is moderated by the EE relationship. We discuss the theoretical contributions of our results and explain key managerial implications for organisations hoping to reap the benefits of ESM in a sustainable way.
1 INTRODUCTION
Knowledge is the most important resource for organisational success because it encapsulates valuable intangible assets and unique organisational routines that are difficult to imitate (Argote & Fahrenkopf, 2016; Tsoukas, 2009), and because it generates creative solutions to problems (Davison et al., 2013; Grant, 1996). Therefore, knowledge contributions—individual acts of voluntarily providing knowledge to others (Ma & Agarwal, 2007)—are vitally important to organisations (Qureshi et al., 2018; Tams, Dulipovici, et al., 2020; Wasko & Faraj, 2005).
This is why the use of Enterprise Social Media (ESM) has skyrocketed over the last decade (Leonardi & Neeley, 2017). ESM consists of ‘computer-mediated tools of the Web 2.0 generation that make it possible for anyone to create, circulate, share, and exchange information in a variety of formats and with multiple communities’ (Leonardi & Vaast, 2017). These tools offer modern organisations new ways to leverage the valuable knowledge possessed by their members (Beck et al., 2014; Engelbrecht et al., 2019; Leonardi, 2017). Many ESM platforms—including Slack, Jive, Salesforce Chatter, and Microsoft Teams—have been widely adopted and are being extensively used at organisations all around the world (Boukef & Charki, 2019; Leonardi & Neeley, 2017). In 2022, Teams alone had over 270 million monthly users who were actively contributing their knowledge on the platform (ZDNET, 2022). ESM has become an essential component of the knowledge strategies adopted by numerous firms (Leonardi, 2017). This is no surprise given that its capacity to improve performance (Kuegler et al., 2015; Ma et al., 2022), productivity (Wu, 2013; Yang et al., 2021), and innovation (Bodhi, Luqman, et al., 2022; Leonardi, 2014) is well documented.
Most of the benefits of using ESM are connected to the unique functionalities that it offers, functionalities that were less prevalent or even non-existent in prior organisational technologies (Leonardi & Vaast, 2017). For example, new features that allow users to display information about contributors and about the knowledge that they possess (Engelbrecht et al., 2019), and ones that allow users to manage social networks and to obtain easy access to digital knowledge (Kane, 2017), are highly beneficial. Other new features of ESM provide users the opportunity to engage in work-related and nonwork-related interactions simultaneously (Neeley & Leonardi, 2018). In work-related interactions, employees use ESM to interact with their colleagues about topics directly connected to their duties and responsibilities in their jobs (Kuegler et al., 2015). Among other things, these interactions focus on problem-solving, incident management, project coordination, and general productivity (Wu, 2013). In nonwork-related interactions, employees communicate with their colleagues about topics that have no connection to their jobs, ones that are mostly related to hedonic activities that they enjoy outside of work (Howe et al., 2022). These interactions can have to do with hobbies and leisure pursuits such as sports, cooking, travel, arts, and music (Huang et al., 2015). This clearly differentiates ESM from other types of IS that users tend to reserve for work-related interactions.
However, although researchers have examined the coexistence of these two types of interactions on ESM, they have arrived at very different conclusions about how nonwork-related interactions affect the work domain. On the one hand, there are studies that highlight the negative impact of nonwork-related interactions, stressing that they can disrupt work (Gibbs et al., 2015), cause a symbolic divide between the work and the nonwork domain (Treem, Dailey, Pierce, & Leonardi, 2015), increase perceptions of overload (Chen & Wei, 2019), and reduce employee performance (Zhang et al., 2023). On the other hand, there are studies that highlight the benefits of nonwork-related interactions on ESM, stressing that they can trigger employee curiosity about work-related content (Neeley & Leonardi, 2018) and increase the readership of work-related posts (Huang et al., 2015). Given this tendency to highlight either the negative or the positive impact of nonwork interactions, we believe that it is time to investigate the connection between work-related and nonwork-related interactions on ESM more thoroughly by focusing on the issue of knowledge contributions on ESM. Hence the two following research questions: (1) How are nonwork-related knowledge contributions on ESM connected to work-related knowledge ones? and (2) What are the factors that condition the relationship between these two types of knowledge contributions?
We believe that answering these research questions is important for two main reasons. First, given that ESM is only useful to organisations if their employees use it to contribute work-related knowledge that can be shared with their colleagues to the benefit of these colleagues (Leonardi, 2017) and that of the organisation as a whole (Kuegler et al., 2015), it is essential to identify the factors that condition this knowledge contribution behaviour. Once these factors have been identified, it will be possible to promote the conditions that favour work-related knowledge contributions and to remedy the ones that hinder it. Second, the use of ESM for nonwork-related purposes is more and more prevalent in organisational life (Leonardi & Vaast, 2017), and to attract and retain talent, employers are increasingly encouraged to be flexible about this nonwork-related use of ESM and even to provide their employees the resources necessary to communicate about their nonwork-related interests while they are on the job (Howe et al., 2022). This means that it is important to understand the effects of using ESM for nonwork-related purposes in work contexts, for obtaining a more complete picture of the impact of ESM use on organisations will improve understanding of how it affects fundamental employee behaviours such as contributing work-related knowledge. This is particularly true given that IT is increasingly used for nonwork-related purposes (Jiang et al., 2023) and that IT use tends to lead to frequent spillover between the work and the nonwork domains (Benlian, 2020).
Drawing on social exchange theory (Blau, 1964; Gouldner, 1960) and on the associative–propositional evaluation model (Gawronski & Bodenhausen, 2006, 2014), we propose a conceptual model that theorises the relationship between work-related and nonwork-related knowledge contributions. Given that knowledge contributions are a typical form of social exchange, social exchange theory (SET) provides an ideal conceptual framework for investigating the issues identified in our research questions. According to SET, people engage in social exchanges after they carefully assess the costs and benefits of these activities. That is, when employees make knowledge contributions, they share precious resources with others because they believe that the rewards obtained through this effort exceed the losses incurred (Rode, 2016). Furthermore, reciprocity—the belief that if one engages in an activity providing benefits to other people, then these people will provide one with similar benefits in the future—is a key premise of SET (Cook, 1977; Cook et al., 2013). Accordingly, when employees make work-related knowledge contributions that benefit the organisation they work for, they do so because of the future rewards they expect to receive, or simply because of their need to reciprocate the benefits they have already obtained from this organisation (e.g., using the ESM platform to engage in a form of play at work).
Building on these ideas, we argue that a knowledge contribution on ESM is not only a social exchange that occurs between one individual employee and a group of colleagues, but also one that involves the organisation where it is enacted and where these employees work. In the case of nonwork-related knowledge contributions on ESM, this entails that these contributions can positively influence work-related ones because of the way that employees interpret the organisation's decision to create a digital space where, in addition to communicating about work-related issues, they can enjoy communicating about their nonwork-related interests. We then argue that the relationship between these two types of knowledge contributions is moderated by the EE relationship—the relationship between employees and their employer (organisation)—which we theorise in terms of perceived organisational support (POS) and employee psychological safety (EPS). We maintain that this relationship is strengthened when employees perceive that their organisation supports them and when they perceive that their work environment is psychologically safe. The higher the perception of organisational support and psychological safety in the work environment is, the stronger the EE relationship is, and the stronger the desire to reciprocate is. To test our hypotheses, we collected survey data from 269 employees of a French e-commerce firm with an ESM platform. The results of the data analysis support the hypothesis that nonwork-related knowledge contributions can positively influence work-related ones and the hypothesis that the EE relationship plays a moderating role in the relationship between these two forms of knowledge contributions.
Our study extends research in three ways. First, it proposes the concept of nonwork-related knowledge contributions to take account of a specific type of behaviour that occurs on ESM. This concept moves beyond the general IT use constructs that researchers have previously developed to examine online behaviours such as blogging (Lu et al., 2015), posting (Huang et al., 2015), and interacting (Neeley & Leonardi, 2018). Indeed, prior studies have seldom focused on the nonwork-related use of ESM as a type of knowledge contribution behaviour, for researchers usually see knowledge contribution behaviour as restricted to the work domain (see Appendix A for an overview of the literature on this theme). We argue that knowledge contributions on ESM can also be about nonwork-related topics and that when they are about such topics, they consist of a distinct type of knowledge behaviour that merits research attention. Second, our study also extends research by theorising and empirically validating the assertion that although they may appear to be disconnected, work-related knowledge contributions on ESM and nonwork-related ones are interdependent behaviours. Third, it extends prior qualitative research on the ways that organisational contexts influence the relationship between the work and the nonwork domain (Neeley & Leonardi, 2018). It does so by theorising the employee–employer (EE) relationship and by showing that employee perceptions of organisational contexts can significantly impact the relationship between work-related and nonwork-related uses of ESM in these contexts.
The rest of the article is organised as follows. In the next section, we provide an overview of ESM and its work-related and nonwork-related uses. We then discuss our study's theoretical underpinning in order to develop a conceptual framework for examining work-related and nonwork-related knowledge contributions on ESM and the relationship between them. After that, we present our research model, our main hypotheses, and the methodological steps taken to validate them. In the final section, we discuss the theoretical and the practical implications of our findings and highlight promising avenues for future research.
2 THEORETICAL BACKGROUND
2.1 Enterprise social media and work-related knowledge contributions
There is no question that knowledge sharing within firms is an important capability and a major driver of competitive advantage (Grant, 1996). Work-related knowledge contributions can bolster strategy making and help facilitate strategy execution (Jarzabkowski, 2004; Neeley & Leonardi, 2018). However, their capacity to do so depends on whether employees are willing to share their knowledge and on whether the knowledge shared can be diffused across the organisation and routinized (Grant, 1996; Nonaka & Takeuchi, 1995; Yang & Chen, 2007). In this respect, one of the most important obstacles is the divide that often exists between employees who are willing to share knowledge and those who need to apply it. For knowledge to be beneficial to the organisation, those who possess it must be identified and relevant contextual information to help decode and interpret it must be available and easily acquired (Bailey et al., 2012; Brown & Duguid, 2001; Nicolini et al., 2012). This can be difficult because employees do not always know who among their colleagues possesses relevant knowledge (Szulanski, 2003).
ESM offers a possible means to surmount such obstacles because it makes accessible detailed information about organisational members and about the knowledge possessed by them (Leonardi & Meyer, 2015). By viewing profiles on ESM and scrolling through prior conversational threads, employees can easily learn about other colleagues and explore the intricacies of their knowledge and the contexts of its application (Leonardi, 2014). ESM platforms provide greater communication visibility (Engelbrecht et al., 2019), they improve productivity (Wu, 2013), and they increase innovation (Leonardi, 2014). Consequently, many organisations have invested in ESM as a new form of knowledge management system that gives employees greater access to valued work-related knowledge (Kane, 2017).
Work-related knowledge contributions are a fundamental prerequisite for desirable organisational outcomes (Kankanhalli et al., 2005; Qureshi et al., 2018; Wasko & Faraj, 2005), and they are the main reason why organisations consider ESM platforms highly useful (Leonardi, 2017; Li, 2015). Those who engage in knowledge contribution behaviour do so for many reasons. For example, in his study of a large multinational high-tech firm, Rode (2016) found that employees contribute work-related knowledge on ESM to build their reputations as experts, to benefit from expected reciprocal behaviour, and to satisfy their need for knowledge self-efficacy. In their study examining a large business software development firm, Chen et al. (2020) found that communications visibility based on message transparency1 and network translucency2 strongly influences work-related knowledge contributions on ESM.
2.2 Enterprise social media and nonwork-related knowledge contributions
Like all organisational IS, ESM offers numerous affordances (action possibilities) to users who, depending on their abilities and motivations, can have various different perceptions of them or even fail to perceive them at all (Engelbrecht et al., 2019). However, the affordances offered by ESM are not limited to the work domain. In fact, as mentioned earlier, ESM is closely tied to the nonwork domain because it allows employees to share knowledge about personal interests that have no connection to their jobs but that they are passionate about (Howe et al., 2022). Indeed, many employees now use ESM to communicate with online groups about topics related to their personal interests just as often as they use it to communicate with groups of colleagues about topics related to their work activities (Ma et al., 2022; Van Osch et al., 2023).
It is important to stress that ESM consists of social media platforms, for the use of such platforms initially became popular outside the business world and only became part of organisational life much later (Leonardi & Vaast, 2017). Most employees started using social media in their personal lives long before their employers became aware of its potential organisational benefits as a communications tool (Gonzalez et al., 2015; Sinclaire & Vogus, 2011). This initial nonwork-related interest in social media platforms sets them apart from most other modern technologies, whose use in the nonwork-related public sphere has usually only occurred after an initial work-related interest in their benefits has sparked their development. This explains why negative attitudes often taint the way that ESM and the affordances that it offers are perceived (Fieseler et al., 2015; Koch et al., 2013; Treem et al., 2015). Indeed, it is not uncommon for the nonwork-related uses of ESM to be perceived as an adverse side effect of its adoption by organisations or as a form of IT misuse (Song et al., 2019) that should be avoided because it can hamper work-related activities (Neeley & Leonardi, 2018). Thus some researchers argue that the nonwork-related use of ESM can increase gossip at work (Leonardi et al., 2013), disrupt work flow (Gibbs et al., 2015), and even tarnish employees' reputations because of the excessive socialisation that it provokes (Neeley & Leonardi, 2018). The claim that it causes information overload (Chen & Wei, 2019) and the claim that it leads to reduced employee collaboration (Zhang et al., 2023) are among other frequently cited negative criticisms of the use of ESM for nonwork-related purposes.
Despite these criticisms, ESM offers employees new ways to express who they are as people in communications that transcend the work domain (Estell et al., 2021; Leonardi & Vaast, 2017). By using ESM for nonwork-related purposes, employees go beyond posting (Huang et al., 2015), participating (Neeley & Leonardi, 2018), and blogging (Lu et al., 2015), for they contribute a distinct category of knowledge—nonwork-related knowledge—that can help colleagues solve personal problems, obtain new insights about topics of interest, or simply feel more connected to others. This nonwork-related use of ESM has been shown to increase the affective organisational commitment of employees even if it does not directly improve their work efficiency (Luo et al., 2018).
ESM is now an essential communications component of many organisations (Karoui et al., 2015). Seen as an excellent means of supporting collaboration (Kwahk & Park, 2016) and shared cognition (Leonardi, 2018), it is adopted by organisations to improve the assimilation (Engelbrecht et al., 2019), the management (Kane, 2017), and the exchange of knowledge (Beck et al., 2014). At the same time, it is also a gathering space where employees can connect with one another to discuss topics and concerns that are not related to work activities. Indeed, ESM gives employees an ideal venue to voice their ideas and opinions, one that is much more impactful than traditional organisational channels (Estell et al., 2021). However, although the interaction between the work-related and the nonwork-related use of ESM has been acknowledged by researchers (Neeley & Leonardi, 2018), the relationship between them has not been sufficiently examined. As Appendix A shows, most research has focused on work-related knowledge contributions on ESM, even if some studies have alluded to possible links between work-related interactions and nonwork-related knowledge contributions on ESM (Neeley & Leonardi, 2018). The next section builds on SET and on the APE model to develop a conceptual framework to help explain the relationship between work-related and nonwork-related knowledge contributions on ESM.
2.3 Social exchange theory
According to SET, a social exchange is a voluntary transfer of resources between two or more social actors (Cook, 1977). The exchange and the ensuing relationships between the parties involved are shaped by assessments of the costs and benefits associated with the exchange process (Blau, 1964). The main premise of SET is that people strive to maximise benefits and minimise costs when they engage in social activities. In other words, people engage in a social exchange because they believe that the potential benefits of the exchange exceed any potential losses. These costs and benefits can be material and tangible, or they can be emotional and psychological. In social exchanges, opportunity costs are also taken into consideration. These are costs associated with sacrificed benefits that could have been redeemed if alternative courses of action had been adopted (Molm, 1990).
The costs and benefits of a social exchange are not always immediate. Because of reciprocity norms, benefits often materialise at a later date. Indeed, the actors in a social exchange usually anticipate or expect some form of future reciprocal action (Cook, 1977; Cook et al., 2013). According to SET, people are well aware of the requirement to reciprocate the support or assistance that is received from others (Blau, 1964); therefore, they engage in social exchanges if others have acted favourably toward them in the past or if they expect them to do so in the future. In short, reciprocity in social exchanges depends on the belief that if one engages in an exchange that provides benefits to other people, then these people will provide one with similar benefits when a suitable opportunity to do so arises in the future. Since this notion of expected reciprocity implies that the outcomes of social exchanges are normally positive and equally advantageous to all the parties involved, it is essential to the stability of social exchange relationships (Emerson, 1976). If people did not embrace the notion of expected reciprocity, then social exchanges would become unstable—perhaps even inconceivable.
2.4 Knowledge contributions on ESM as social exchanges
Since they involve voluntary interactions between two or more social actors, knowledge contributions on ESM are social exchanges (Cook, 1977). As such, these knowledge contributions and the relationships that transpire between the parties involved are driven by cost–benefit considerations. That is, employees will only use ESM if the hedonic and utilitarian benefits that they draw from this use outweigh its costs. This interpretation of ESM use is consistent with the findings of most studies on social media (Beck et al., 2014; Wenninger et al., 2019).
SET is typically used to examine dyadic exchanges such as the communicational and transactional ones that occur between employees and members of their professional social networks (Beck et al., 2014; Matook et al., 2015). However, we argue that in the analysis of interactions on ESM, it is also necessary to use SET to conceptualise the organisation where employees work by treating it as an entity that has a participating role in the social exchanges that occur on ESM (Cropanzano et al., 2017). This is because, in addition to the individual contributors who are generally the focus of social exchanges in traditional social media settings, social exchanges in ESM contexts are bounded by the organisational environment that makes them possible in the first place. Even if ESM use by individual employees is an inherently voluntary activity, the initial acquisition and the continued provision and maintenance of an ESM platform are generally based on organisational decisions (Kane, 2015; Leonardi et al., 2013). Consequently, the analysis of social exchanges on ESM must take into consideration the role of the organisation and cannot be restricted to the individual employees engaged in these exchanges. Since it has made possible this type of social exchange among its employees by implementing and maintaining an ESM platform, the organisation must also be considered an actor in these social exchanges. Finally, this means that in this type of social exchange, the way employees view the organisation they work for and how they assess their relationship with it can be expected to influence the reciprocal behaviour that they adopt (Eisenberger et al., 2019).
3 THEORETICAL DEVELOPMENT
In organisational behaviour studies that theorise social exchanges, three distinct components are often conceptualised (Cropanzano et al., 2017). The first component is the initiating behaviour of an actor that triggers the exchange—frequently a behaviour that positively or negatively affects a target individual. The second one is the social relationship that transpires from the exchange and that determines the quality of the rapport between the parties involved. The third one is the reciprocating behaviour that the target individual adopts in response to the initiating actor and that determines the quality of the social exchange relationship as a whole.
If this conceptual framework is used to analyse social exchanges on ESM, then the organisational decision to implement an ESM platform can be theorised as the initiating behaviour that triggers them. Since, as discussed earlier, ESM has important nonwork-related uses (Fieseler et al., 2015; Luo et al., 2018; Treem et al., 2015), this decision can be interpreted as an initiative to introduce into the work environment a space that can be used, not only for serious work-related communications, but also for enjoyable nonwork-related communications that are a form of play at work (Petelczyc et al., 2018). In other words, it can be interpreted as an initiative that offers employees the opportunity to enjoy discussing their personal interests with colleagues while on the job (Howe et al., 2022). This interpretation is plausible because the use of ESM for nonwork-related purposes has the same defining characteristics as most other forms of play at work (see the review of Petelczyc et al., 2018): (1) its main goal is fun and amusement; (2) it revolves around freely chosen activities that employees are enthusiastic about; and (3) it involves highly interactive social exchanges (Van Vleet & Feeney, 2015). This makes it clear that employees reap personal gains from their use of ESM for nonwork-related purposes, personal gains connected to the pleasurable experience that the platform offers them. If these gains are interpreted in accordance with the reciprocity norms highlighted by SET, then the employees who reap them can be expected to give back to their organisation by engaging in work-related activities on ESM, for social exchanges generally involve reciprocation when their outcomes are positive and equally advantageous to all the parties involved (Emerson, 1976).
Although the organisational decision to introduce an ESM platform offers employees the opportunity to engage in a form of play at work (Petelczyc et al., 2018) that is likely to foster the development of high-quality social exchange relationships (Cropanzano et al., 2017), it is but one of many possible organisational decisions, each of which can influence the relationship between the employees and their employer. As depicted in Figure 1, whenever an organisation makes decisions or engages in action, it provides valuable information to its employees that allows them to assess the quality of the social exchange relationship that binds them to the organisation. These assessments can be expected to influence the degree to which employees are willing to engage in reciprocity behaviour to help sustain the social exchange relationship with their employer.
Put simply, when employees believe that their organisation has created a positive digital space where they can enjoy communicating about nonwork-related topics while on the job, they can be expected to reciprocate by engaging in activities such as work-related knowledge contributions that benefit the organisation. However, it is unlikely that this positive relationship between the employees and their employer will be stable across organisational contexts. This is because as an IT artefact, one and the same ESM platform can be experienced and understood very differently by its users depending on the spatial, temporal, discursive, and/or social contexts in which it is embedded (Cheikh-Ammar, 2018; Orlikowski & Iacono, 2001). Most importantly, a positive social exchange relationship between employees and the organisation they work for can only be sustained if these employees continue to engage in reciprocity behaviour, but this behaviour depends on how they assess the work environment and on how they perceive themselves within this environment.
As it can be seen in Figure 2, we conceptualise this employee–employer (EE) relationship as a function of perceived organisational support (POS) and of employee psychological safety (EPS). This choice is motivated by two main considerations: (1) Empirical and theoretical significance in extant literature. POS and EPS are two fundamental beliefs that have been consistently identified in organisational behaviour studies as strong indicators of the quality of the relationship between employer and employees (Edmondson & Lei, 2014; Musenze & Mayende, 2023), making them ideal constructs to include in our theoretical model. In fact, prior research has frequently shown that employees who feel supported and who enjoy high levels of psychological safety in their work environment are more likely to engage in knowledge sharing with their colleagues (Kessel et al., 2012) and to positively contribute to the overall advancement of the organisation (Detert & Burris, 2007). (2) Conceptual relevance to the moderation argument advanced in this study. POS is a general belief that employees hold ‘concerning the extent to which the organization values their contribution and cares about their well-being’ (Eisenberger et al., 1986; Eisenberger & Stinglhamber, 2011; Kurtessis et al., 2017), whereas EPS describes the extent to which ‘one can express oneself without fear of negative consequences to self-image, status, or career’ (Kahn, 1990, p. 708). As such, both constructs consist of global beliefs that employees hold about their work environment; beliefs that can easily influence their commitment to reciprocity behaviours (Cropanzano et al., 2017; Kurtessis et al., 2017). It is so because they are known to be directly related to the attitude of employees at work (Frazier et al., 2017; Newman et al., 2017), and to their level of engagement in extra-role activities such as knowledge contribution (Tams, Dulipovici, et al., 2020). As such, it becomes very plausible that variations in the levels of POS and EPS that employees experience at work would influence the degree to which they feel inclined to compensate their employers for the favourable working conditions they provide (Gouldner, 1960; Mowday et al., 1982).
Using this conceptualization (i.e., POS and EPS), we posit that the EE relationship shapes the spillover from the nonwork-related to the work-related domain on ESM, with employees assessing organisational decisions and actions, then using these assessments to determine the quality of their social exchange relationship with the organisation and adapting their reciprocity behaviour accordingly. There is further conceptual support for the effects of this type of interaction in dual-attitude theories that posit the existence of two inherently different yet complementary cognitive processes affecting human judgement (Chang & Ko, 2016; Petty et al., 2007). In the case of the APE model, an associative and a propositional cognitive process are theorised to explain individual decision-making (Gawronski & Bodenhausen, 2006). According to this model, associative processes are responsible for the automatic activation of associations between concepts stored in a person's memory. These rapid, effortless, and often unconscious processes are triggered by instigating mechanisms such as priming, conditioning, or simple exposure (Gawronski & Bodenhausen, 2006, 2014). As for propositional processes, the model posits that they are more effortful and deliberate because they involve the use of conscious thinking and logical reasoning to assess possible discrepancies between beliefs and reality (Chang & Ko, 2016; Petty et al., 2007).
For example, it is plausible to assume that persons who associate electric cars with environmental benefits will tend to say, more or less automatically, that they prefer these cars to traditional gasoline-powered ones, if they are asked to compare these two types of vehicles. However, it is important to consider what will happen if these persons are presented with new information about other positive aspects of electric cars, aspects such as the low cost of vehicle maintenance and the wide availability of public charging stations, or if they are presented with new information about the negative aspects of electric cars, aspects such as the energy-intensive manufacturing process required to build these cars and the environmental impact of battery production that depends on the extraction of rare minerals. According to the APE model, this new information will trigger propositional processes when these persons attempt to assess their initially held beliefs about electrical cars while taking into consideration these additional facts that they have learned about them. In other words, they will engage in critical thinking, evaluating the evidential plausibility of the new information and assessing what they already know and believe about electrical cars, environmental sustainability, and the automobile industry in the light of it. This will either strengthen their positive association of electrical cars or weaken it.
Thus the APE model posits that human judgement involves the interplay of associative and propositional processes. According to the model, both processes can influence human behaviour, but depending on the context, their respective impacts on human behaviour can be subject to important variations. Associative processes are dominant in familiar contexts that require limited cognitive effort, and they usually lead to more or less automatic responses that are based on preexisting associations. Propositional processes come into play in contexts where new information emerges that requires more careful assessment, and they tend to foster more thoughtful decision-making (Gawronski & Bodenhausen, 2006, 2014).
Coupled with our research model, the APE model provides us with the theoretical underpinning necessary to argue that the positive associations that employees make between social media and hedonic pursuits increase the likelihood that they will view the organisational decision to adopt an ESM platform as a positive initiative that allows them to engage in a form of play at work (Petelczyc et al., 2018)—communicating about their nonwork-related personal interests while on the job. It is also plausible to assume that this interpretation of the organisational decision to implement an ESM platform will lead employees to evaluate the social exchange relationship that they have with their employer positively and to reciprocate by engaging in work-related activities. However, we also argue that this reciprocity behaviour and the sustainability of the social exchange relationship between the organisation and its employees are largely dependent on additional information that the latter retrieve about the organisational environment. This additional information activates propositional reasoning processes that involve careful analysis and deliberation to assess the organisational environment, and this assessment either accentuates or diminishes the reciprocity behaviour of employees accordingly. In other words, just as the new information in the example about electric cars triggers critical thinking about the positive association of these cars with environmental benefits, perceived organisational support (Kurtessis et al., 2017) and the organisational climate of psychological safety (Edmondson, 2018) offer important contextual cues that lead to the strengthening or the weakening of preexisting positive associations that employees make between ESM and hedonic pursuits, and thus to the accentuation or the diminution of their reciprocity behaviour. Gawronski and Bodenhausen (2006) refer to these evaluations that help determine the validity of existing associations as syllogistic inferences.
3.1 Research model and hypotheses
The theoretical framework presented in Figure 1 offers a high-level conceptualization of the social exchange process that unfolds following the organisational adoption of an ESM platform. However, this conceptualization does not explicitly identify the key factors that shape this social exchange process and the way that it impacts the use of ESM by employees to contribute knowledge. To take account of these factors, we introduce the variance model presented in Figure 2.
This model highlights two types of knowledge contribution behaviours on ESM—one related to the work domain and one related to the nonwork domain. The relationship between these two interconnected behaviours depends on the concern and encouragement that an organisation expresses to its employees, and on the degree of freedom (Eisenberger et al., 2019) and risk-taking that its employees perceive as possible for them in organisational contexts (Edmondson, 2004). In addition, the model posits that POS and EPS are two fundamental aspects of the EE relationship that have the combined effect of either strengthening or weakening the desire that employees have to engage in reciprocity behaviour. When there is a significant level of reciprocity behaviour, social exchange outcomes on ESM can be expected to be positive and equally advantageous to all the parties involved (Emerson, 1976). Table 1 presents all the constructs in the research model in Figure 2 along with their conceptual definitions. The relationships specified in the model are discussed in the next section.
| Construct | Definition | Reference |
|---|---|---|
| Work-related knowledge contribution | The voluntary act of providing useful work-related knowledge to colleagues on ESM | Adapted from Ma and Agarwal (2007) |
| Nonwork-related knowledge contribution | The voluntary act of providing useful nonwork-related knowledge to colleagues on ESM | Adapted from Ma and Agarwal (2007) |
| Perceived organisational support | The general belief of employees that their work organisation values their contribution and cares about their well-being | Eisenberger et al. (1986) |
| Employee psychological safety | The perception that employees have that they can be themselves in organisational contexts without fear of negative consequences for their self-image, status, or career | Kahn (1990) |
3.1.1 From nonwork-related to work-related knowledge contributions
The relationship between work-related and nonwork-related knowledge contributions on ESM involves two mechanisms. The first one is connected to the social exchange considerations that shape the relationship between employees and the organisation that they work for. The second one is connected to the social exchange considerations that shape the relationship between employees and their colleagues in the social network that develops on the ESM platform at the organisation they work for.
To understand the first mechanism, it is important to recall that according to SET, employees can be expected to reciprocate if they believe that their organisation has acted favourably in their regard. It is also important to recall that employees are likely to view the organisational decision to implement ESM as a positive initiative that creates a space where they can enjoy communicating about their nonwork-related personal interests while on the job (Petelczyc et al., 2018). If their positive assessment of the organisation's decision to adopt an ESM platform is interpreted in line with SET, then we should expect employees to feel obliged to reciprocate by engaging in work-related activities and in particular by making work-related knowledge contributions on ESM. Moreover, as employees continue using the ESM platform for nonwork-related purposes, they develop technological frameworks that shape their perceptions of the affordances that the platform offers and the values that these affordances project (Leidner et al., 2018), and these perceptions come to shape the ways that they appropriate and use the platform (Neeley & Leonardi, 2018; Treem et al., 2015). Thus, as ever-recurring perceived benefits, the opportunities to make nonwork-related knowledge contributions on ESM reinforce the reciprocity behaviour of employees, who increasingly make work-related knowledge contributions that have instrumental benefits for the organisation.
The second mechanism also has to do with reciprocity norms, but this time it is a question of reciprocity norms that bind colleagues together as they initiate nonwork-related knowledge contributions that have an impact on work-related ones. Nonwork-related knowledge contributions on ESM tend to arouse employees' curiosity and draw them back to the platform. In addition, when employees provide information about nonwork-related topics, they learn about their colleagues in the ESM network and these colleagues learn about them in turn. Just as in work-related communications, this type of exchange enhances message transparency by making the content of the knowledge shared accessible to other colleagues, and it improves network transparency by making employees' social networks apparent to colleagues (Leonardi, 2015). In fact, the visibility that ESM gives users is one of its key properties, which means that employees who possess specific types of knowledge become more easily identifiable (Van Osch & Steinfield, 2018). This facilitates information requests, for on ESM they can be immediately directed to the appropriate persons. As a result, employees no longer have to be afraid of appearing less knowledgeable because of misdirected questions about important topics that they know little or nothing about (Neeley & Leonardi, 2018). In addition, the nonwork-related content of communications on ESM often serves as a gateway to the discovery of useful work-related content because employees are drawn to their colleagues' shared content on ESM (Neeley & Leonardi, 2018). Thus, by making employees who possess specific types of knowledge more visible (Van Osch & Steinfield, 2018), ESM stimulates interactions between the members of an organisation (Leonardi, 2015) for the purpose of knowledge sharing (Rode, 2016).
Moreover, as employees learn about one another on ESM through their nonwork-related knowledge contributions, the uncertainty characteristic of knowledge seeking (Beck et al., 2014)—which is often related to the problem of identifying experts and the difficulty of reaching out to unknown individuals—gradually recedes (Leonardi, 2018). This is because knowing more about colleagues increases trust and makes it easier to identify the ones that possess the knowledge needed to resolve problems (Neeley & Leonardi, 2018). When communication is mediated by information technology, trust can be built more easily because information about others can be gathered rapidly and negative perceptions based on an inadequate acquaintance with others are not as difficult to dispel (Jarvenpaa & Leidner, 1999). When employees contribute nonwork-related knowledge on ESM, they reveal themselves and become more transparent to others, and this helps relationships grow and become more intimate (Leidner et al., 2018). Prior research on online communities has shown that intimacy enhances the sense of belonging to a group (Constant et al., 1994; Constant et al., 1996), and there is no doubt that it makes employees more willing to help their colleagues with work-related issues.
In short, when employees contribute nonwork-related knowledge on ESM, they learn about their colleagues and have the opportunity to develop trust in them. As a result, they are more likely to help them by contributing work-related knowledge if the opportunity to do so presents itself. At the same time, they also feel indebted to their organisation for providing them with a digital space for enjoyable communications, and this boosts their willingness to reciprocate by making work-related knowledge contributions that are helpful to their colleagues and to the organisation as a whole. This suggests that it is plausible to assume the following hypothesis:
H1.Nonwork-related knowledge contributions on ESM positively influence work-related knowledge contributions on ESM.
3.1.2 The moderating role of the employee–employer relationship
These two forms of knowledge contributions do not occur in a void, for they are largely dependent on the reigning EE relationship. Moreover, although the two behaviours are interconnected because of the social exchange relationship between employees and the organisation they work for, the quality of this social exchange relationship is not solely determined by the organisational decision to adopt a digital platform that offers a space for enjoyable nonwork-related communications that are a form of play at work (Petelczyc et al., 2018). On the contrary, it is determined by all the decisions and actions of the organisation as well as by the overall work environment that they help to foster. Indeed, employees develop global beliefs about the environment that they work in, beliefs that can determine the extent to which they feel obliged to reciprocate the benefits that they receive from their organisation. In a word, global beliefs about their work environment play an important role in determining whether employees' nonwork-related knowledge contributions on ESM lead to useful work-related ones.
POS and EPS are two important perceptual beliefs that have been frequently examined in organisational behaviour studies, where they have been consistently identified as strong indicators of the quality of the EE relationship (Edmondson & Lei, 2014; Musenze & Mayende, 2023). This suggests that the various decisions and actions of an organisation give employees an overall impression on the basis of which they make propositional evaluations that influence their reciprocating behaviour, for it is reasonable to assume that this behaviour is either accentuated or diminished by the conclusions that they draw in these propositional evaluations. In other words, these conclusions should be understood as moderation effects that have a synergistic pattern, so that positive spillover from nonwork-related to work-related knowledge contributions are more likely when employees feel that it is safe to be themselves in the work environment and when they believe that their employer is concerned about their well-being.
Perceived organisational support
Employees often view actions taken by management and by various organisational agents as actions taken by the organisation as a whole—as if it were a fully unified entity with human agency (Levinson, 1965). This anthropomorphic interpretation of the organisation makes it possible for employees to ascribe to it dispositional traits such as a desire to support them (Eisenberger et al., 1986). According to organisational support theory, employees often hold general beliefs concerning the extent to which the organisation they work for is concerned about their well-being and the extent to which it values their contribution (Eisenberger et al., 1997; Shore & Shore, 1995). This POS is closely linked to the intention to behave benevolently that they attribute to the organisation after they gain experience with its norms, policies, procedures, and actions (Eisenberger et al., 2001).
POS is a fundamental aspect of the employee understanding of the EE relationship (Eisenberger & Stinglhamber, 2011; Han et al., 2019). Thus employees can be expected to take it into account when they consider engaging in action that can impact the organisation they work for. When they believe that this organisation values their input and is concerned about their well-being (Eisenberger et al., 1986), this belief is likely to facilitate the decision to reciprocate positive action that it has undertaken on their behalf. This is because employees are inclined to compensate employers for the favourable working conditions that they provide (Gouldner, 1960; Mowday et al., 1982)—especially when they feel supported (Eisenberger & Stinglhamber, 2011). At an organisation where employees are grateful for the opportunity it provides them to engage in activities that are a form of play at work (Petelczyc et al., 2018) and where they perceive strong organisational support, they are likely to be even more motivated to reciprocate these benefits by engaging in work-related knowledge contributions that are cognitively consistent with them (Gawronski & Bodenhausen, 2006).
However, when employees perceive little organisational support, they tend to doubt the trustworthiness of the organisation and to dislike the work environment. In these conditions, the positive effects of nonwork-related knowledge contributions on work-related ones are diminished because the value of reciprocity behaviour is less salient. In other words, the implementation of a space where they can enjoy communicating about nonwork-related personal interests with colleagues will play a weaker role, and the reciprocity behaviour that transpires from it will have less of an impact. As a result, employees who frequently contribute nonwork-related knowledge on ESM will spend less time and effort contributing work-related knowledge on ESM, and they will be less likely to be highly motivated to do so.
These considerations suggest that when perceived organisational support is high, the benefits of nonwork-related knowledge contributions on ESM are more likely to encourage employees to make work-related ones. This makes it plausible to assume the following hypothesis:
H2.The relationship between nonwork-related and work-related knowledge contributions on ESM is moderated by POS to the extent that this relationship is stronger when employees perceive higher levels of organisational support.
Employee psychological safety
EPS is based on the belief that ‘one can express oneself without fear of negative consequences to self-image, status, or career’ (Kahn, 1990, p. 708). As the perception that there is a low level of interpersonal risk in the work environment, this is a tacit belief that employees hold discreetly, even if they sometimes discuss it explicitly with their colleagues (Edmondson, 1999). Employees who feel psychologically safe in their work environment are generally more receptive to changing (Edmondson, 2004), learning (Carmeli & Gittell, 2009), and innovating (Edmondson, 2018).
Prior research has shown that employees who enjoy high levels of psychological safety in their work environment are also more likely to share knowledge with their colleagues (Kessel et al., 2012), exchange with them (Collins & Smith, 2006), and contribute to organisational improvement (Detert & Burris, 2007). In addition, it has shown that when top management is able to convey to employees that they are safe from professional risk that could threaten their job status or their overall career (Edmondson, 1999), they are more likely to voice their opinions (Turco (2016), to make suggestions, and to present innovative ideas.
From a social exchange perspective, it is clear that employees who feel they are psychologically safe in their work environment and who enjoy using their organisation's ESM platform to communicate about nonwork-related personal interests with their colleagues are more likely to reciprocate by making work-related knowledge contributions. EPS is a function of employee belief in the organisational commitment to offering a safe environment in which employee attitudes and opinions can be expressed when needed (Edmondson, 1999). This important aspect of the EE-relationship can encourage reciprocity behaviour because employees are inclined to recompense employers for the favourable conditions that they provide (Gouldner, 1960; Mowday et al., 1982)—especially when they feel safe in the work environment (Edmondson, 1999; Frazier et al., 2017). Under these conditions, employees often strive to maintain stable social exchanges in which they and their employers are both happy (Eisenberger et al., 2001). This generally means that they give back to the organisation by participating in work-related knowledge contributions.
In contrast, when employees do not feel psychologically safe in the work environment, they are likely to perceive important levels of risk in any action that they take while on the job (Edmondson, 1999). In particular, they are likely to fear the possible negative consequences of using ESM. According to the APE model (Gawronski & Bodenhausen, 2006), when the conclusions reached through propositional reasoning are inconsistent with initial assumptions—as is the case when the perceived level of psychological safety in the work environment is low—this means that they contradict the syllogistic inferences drawn from originally positive associations and dilute the effects of these associations.
This suggests that employees who contribute nonwork-related knowledge on ESM will devote less time and effort to contributing work-related knowledge on ESM. In addition, it also suggests that they are unlikely to be highly motivated to do so, for they will tend to associate negative consequences with their behaviour on ESM, ones such as being perceived as a person who socialises too much at work (Neeley & Leonardi, 2018). Thus we maintain that the benefits of nonwork-related knowledge contributions on ESM are more likely to encourage employees to make work-related knowledge contributions on ESM when the level of employee psychological safety is high. This makes it plausible to assume the following hypothesis:
H3.EPS moderates the relationship between nonwork-related and work-related knowledge contributions on ESM, for this relationship is stronger when employees perceive higher levels of psychological safety in the work environment.
4 METHODS
4.1 Organisational context
Digital Success (DS, a pseudonym) is a medium-sized e-commerce firm in France that had over 500 employees at the time of the data collection for the present study. DS manages an online trading platform where individuals and businesses can post short messages about buying and selling second-hand goods. The users of the platform have free access to its basic functionalities, and they can choose to pay for premium features. For instance, increased visibility is one of the features available for a fee. Sellers who pay for this feature are allowed to post additional high-quality photos to improve the presentation of their products and to increase their chances of selling them.
DS offers an ideal empirical setting for our research. Most of its employees can be considered knowledge workers because making knowledge contributions is a crucial work behaviour that the DS management team appreciates and encourages. The firm operates in a highly competitive market in which there are aggressive competitors such as eBay and Facebook Marketplace. Although DS is the market leader in France, it is constantly challenged by new entrants. To be able to respond effectively to challenges from its competitors and to constantly changing market conditions, DS has adopted a flat organisational structure and operates as an agile enterprise where work-related knowledge contributions by employees are viewed as the key to success.
4.2 ESM context
Slack, a widely used collaboration and messaging platform owned by Salesforce, is the main platform for knowledge contributions at DS; therefore, it is the ESM platform examined in our study. DS encourages its employees to use ESM instead of email for internal knowledge exchanges whenever possible, which explains why Slack is the main technology that they use in their everyday work environment.
It is important to note that although Slack users can share knowledge about numerous work-related and nonwork-related topics, they typically do so as members of various ESM groups that are referred to as Slack channels. These channels and their content are easily identifiable because of the tags that are systematically associated with them. This allows DS employees to avoid confusion when they use Slack to contribute either work-related or nonwork-related knowledge. Figure 3 depicts a Slack screen from a fictional organisational context. Note that the names of the channels—the words following the hashtags—inform users about the topics and the nature of the discussions (e.g., #cool, #cute, and #funny).
A questionnaire survey was the main method that we used to assess our research model empirically. However, before designing this survey, we conducted 20 preliminary interviews with division managers and employees who use Slack at DS. These interviews allowed us to improve our understanding of the use of ESM at DS and to validate our conceptualization of the EE relationship.
As mentioned earlier, Slack contains various channels for employees to contribute knowledge about a wide array of topics. In the interviews, DS employees stated that they use Slack to contribute knowledge about important work-related topics such as coding, debugging, application development, and incident management, but that they also said that they use it for nonwork-related interactions. Indeed, these interactions are an essential part of the organisational culture at DS, with its emphasis on the need to create a sense of togetherness among employees and on the importance of communicating across hierarchical lines. Members of the DS management team explained that they see nonwork-related interactions on Slack as enjoyable exchanges that promote employee well-being and job satisfaction, and that bring everyone in the organisation closer together. As evidence, our key informant—the head of HR development—shared with us a document on organisational culture that she had used when DS took part in the 2018 Great Place To Work (GPTW) competition.3 The document highlights the importance of nonwork-related interactions as fundamental components of organisational culture at DS. It contains statements like this one: ‘At DS, it's natural to talk with your colleagues about your personal interests or about things like ecology’ (Great Place to Work, Culture Document, 2018, DS).
DS employees share much more than basic information about nonwork-related topics. Many of them contribute detailed knowledge about nonwork-related topics that they are passionate about. Often they say that they communicate this knowledge in order to help colleagues benefit from their personal experience. Examples of nonwork-related knowledge include the following: training routines to use when preparing for a half marathon, effective ways to solve DIY challenges, best practices for maximising the benefits of intermittent fasting, and cooking techniques for preparing macarons. The rich content that these employees create can help colleagues develop new insights (Ma & Agarwal, 2007) and learn about issues that interest them. This content sharing has helped bring about the emergence of a new form of knowledge contribution on ESM that goes beyond posting (Huang et al., 2015), blogging (Lu et al., 2015), and commenting (Laitinen & Sivunen, 2020). Moreover, these nonwork-related knowledge contributions occur primarily on Slack since the intensive nature of the work performed at DS makes efficiency an essential issue. In other words, employees are encouraged to avoid using multiple technologies in parallel, particularly if these technologies support similar tasks. This means that employees do not use private social media platforms such as Facebook to share nonwork-related knowledge with their colleagues.
The interviews also revealed that DS is concerned about EPS and POS. This concern is clearly expressed in the following excerpt from the firm's 2018 GPTW report: ‘DS is committed to the well-being of its employees. Various policies have been introduced, tested, renewed, or discontinued to meet the expectations of our employees, policies designed to improve daily life at work, to make every day a learning experience, to promote leadership excellence, and to acknowledge the contributions of employees’. This concern was also confirmed in the interviews with DS employees, many of whom reported that they are free to contribute to discussions on the platform in any way that they see fit, without fear of being judged or penalised. Indeed, management does not attempt to monitor the content shared on the platform. As one high-ranking manager put it: ‘Anyone can post and react to posts on the ESM platform, and given our culture of transparency, we don't monitor anything. We've never had to delete anything’. This same viewpoint was expressed by the HR manager: ‘We've given employees the freedom to use ESM and to adapt their use of it to any topic they want, whether or not it's work-related’.
4.3 Survey data collection
We designed an online survey for the purpose of empirically assessing the research model presented in Figure 2. In the spring of 2019, the survey was distributed to all DS employees by email via Qualtrics, a popular online survey platform. Three reminders were later sent to the pool of respondents.
For the 510 emails sent, 304 responses were received, yielding a 59.6% response rate. However, 35 responses were deemed incomplete and had to be eliminated, yielding 269 usable questionnaires. Table 2 presents the respondents' demographic characteristics. We checked for nonresponse bias using the procedure proposed by Armstrong and Overton (1977). Thus we created four quartiles of respondents and compared the first group (25% of respondents) with the last one (25% of respondents). We used t tests to compare these groups, and the results revealed no significant differences between respondents and non-respondents with respect to key demographic characteristics.
| Category | Frequency | Ratio | |
|---|---|---|---|
| Gender | Male | 150 | 55.8 |
| Female | 119 | 44.2 | |
| Total | 269 | 100 | |
| Age | <31 | 96 | 35.7 |
| 31–39 | 124 | 46.1 | |
| 40–49 | 45 | 16.7 | |
| >49 | 4 | 1.5 | |
| Total | 269 | 100 | |
| Education | High school | 41 | 15.2 |
| Bachelor's degree | 31 | 11.6 | |
| Master's degree | 164 | 61 | |
| Others | 33 | 12.2 | |
| Total | 269 | 100 | |
| Rank | Regular employee | 201 | 74.7 |
| Manager | 68 | 25.3 | |
| Total | 269 | 100 | |
| Job field | Administration | 60 | 22.3 |
| Marketing | 35 | 13.0 | |
| Sales | 59 | 21.9 | |
| Tech | 115 | 42.8 | |
| Total | 269 | 100 |
4.4 Measurement
We measured all items using a 5-point Likert scale ranging from ‘strongly disagree’ (1) to ‘strongly agree’ (5). These measurement items are presented in Appendix B. The items measuring employee psychological safety (EPS), perceived organisational support (POS), and the two forms of knowledge contributions were adapted from the literature to fit the study context. We measured (EPS) by way of a widely used scale developed by Burris et al. (2008). We measured POS by way of a widely used scale that we adapted from Rhoades et al. (2001). For the two forms of knowledge contributions, the items were taken from Ma and Agarwal (Ma & Agarwal, 2007). However, to help respondents distinguish between the two forms, we explicitly informed them of the knowledge domain that the questions were designed to assess. That is, before responding to the set of questions for each form of knowledge contribution, the respondents had to read a directive indicating which form would be asked about. For the questions on work-related knowledge contributions, the directive was this: ‘For your professional use of work-oriented channels on Slack, please indicate the extent to which you agree with the following statements’. For the questions on nonwork-related knowledge contributions, the directive was this: ‘For your personal use of Slack to communicate about nonwork-related topics such as cooking or sports, please indicate the extent to which you agree with the following statements’.
We controlled for the effects of the following variables: gender, rank, education level, job field, and organisational tenure (number of years working for DS). We also controlled for the effect of organisational citizenship behaviour (OCB) toward colleagues, OCB being ‘the set of positive workplace behaviours that are distinct from the employee's work tasks and that support organization members and/or the work environment’ (Carpenter et al., 2014, p. 547). The effect of OCB on knowledge sharing is well documented (Sadegh, 2015); therefore, it is important to examine its possible role in work-related and nonwork-related knowledge contributions on ESM. We measured OCB toward colleagues using five items adapted from Ong et al. (2018). Table 3 presents the descriptive statistics and the correlations between the variables.
| Mean | SD | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1. WKC | 3.59 | 0.89 | 1 | |||||||||
| 2. NWKC | 2.85 | 1.03 | 0.423** | 1 | ||||||||
| 3. OCB | 3.97 | 0.65 | 0.213** | 0.131** | 1 | |||||||
| 4. EPS | 3.85 | 0.90 | 0.149* | −0.054 | 0.194* | 1 | ||||||
| 5. POS | 3.57 | 0.72 | 0.076 | −0.003 | 0.180** | 0.586** | 1 | |||||
| 6. Education | 3.81 | 1.07 | −0.166** | −0.068 | 0.053 | −0.043 | −0.002 | 1 | ||||
| 7. Age | 33.72 | 6.34 | 0.035 | −0.031 | 0.120* | 0.093 | 0.090 | −0.012 | 1 | |||
| 8. Org. tenure | 2.89 | 2.32 | −0.037 | −0.043 | −0.014 | 0.032 | 0.062 | 0.138* | 0.373** | 1 | ||
| 9. Gender | 0.44 | 0.49 | −0.143* | −0.129* | −0.063 | −0.020 | 0.057 | 0.081 | −0.117 | 0.024 | 1 | |
| 10. Rank | 0.25 | 0.43 | −0.067 | −0.027 | −0.006 | 0.121* | 0.188** | −0.005 | 0.335** | 0.270** | 0.050 | 1 |
| 11. Job field | 2.85 | 1.19 | 0.319** | 0.205** | −0.031 | −0.004 | −0.131* | −0.148* | 0.028 | −0.212** | −0.359** | −0.150* |
- * Correlation is significant at the 0.05 level.
- ** Correlation is significant at the 0.01 level.
4.5 Results
4.5.1 Construct reliability and validity
We evaluated the quality of our survey by assessing the reliability as well as the convergent and discriminant validity of the measurement items. We used SPSS version 27 to calculate all the statistics and all the examined measures of Cronbach's Alpha coefficient and composite reliability. Cronbach's Alpha coefficient and composite reliability were both above the recommended threshold of 0.7 (see Table 4), indicating satisfactory internal consistency (Nunnally, 1978). Moreover, our scales exhibited good discriminant validity since the square root of the average variance extracted (AVE) for each construct was higher than the inter-construct correlations (Table 5). The AVEs for all the constructs were above 0.5 (with the related values ranging between 0.532 and 0.821), indicating satisfactory convergent and discriminant validity. In line with the recommendation of Henseler et al. (2015), we used the heterotrait-monotrait (HTMT) ratio of correlations to do a further assessment of discriminant validity. Our analyses showed that all the HTMT values were below the threshold value of 0.85, which provided further confirmation of discriminant validity (see Table 6).
| Constructs | Cronbach's alpha (CA) | Composite reliability (CR) | Average variance extracted (AVE) |
|---|---|---|---|
| Work-Related Knowledge Contribution (WKC) | 0.868 | 0.910 | 0.718 |
| Nonwork-Related Knowledge Contribution (NWKC) | 0.910 | 0.937 | 0.789 |
| Organisational Citizenship Behaviour (OCB) | 0.724 | 0.819 | 0.532 |
| Perceived Organisational Support (POS) | 0.827 | 0.875 | 0.547 |
| Employee Psychological Safety (EPS) | 0.890 | 0.932 | 0.821 |
| Constructs | WKC | NWKC | OCB | EPS | POS |
|---|---|---|---|---|---|
| Work-Related Knowledge Contribution (WKC) | 0.847 | ||||
| Nonwork-Related Knowledge Contribution (NWKC) | 0.423 | 0.888 | |||
| Organisational Citizenship Behaviour (OCB) | 0.213 | 0.131 | 0.729 | ||
| Employee Psychological Safety (EPS) | 0.149 | −0.054 | 0.194 | 0.906 | |
| Perceived Organisational Support (POS) | 0.076 | −0.003 | 0.180 | 0.586 | 0.740 |
- Note: The square root of the average variance extracted (AVE) is shown on the diagonal of the correlation matrix, and inter-construct correlations are shown off the diagonal.
| Constructs | WKC | NWKC | OCB | EPS |
|---|---|---|---|---|
| Nonwork-Related Knowledge Contribution (NWKC) | 0.469 | |||
| Organisational Citizenship Behaviour (OCB) | 0.294 | 0.187 | ||
| Employee Psychological Safety (EPS) | 0.180 | 0.067 | 0.246 | |
| Perceived Organisational Support (POS) | 0.105 | 0.055 | 0.236 | 0.682 |
In addition, we ran a confirmatory factor analysis to assess model fit. This analysis showed that the five-factor model (NWKC, WKC, OCB, EPS, POS) presented a good fit with our data, with the chi-square over the degree of freedom (χ2/df) of 1.58 (< the recommended threshold of 3), a comparative fit index (CFI) of 0.963, and an incremental fit index (IFI) of 0.964. As explained in the literature (Magni et al., 2018), when CFI and IFI are close to 1, this indicates that a model has good fit. Moreover, the root mean square error of approximation (RMSEA) had a value of 0.047 (< a threshold of.08), which also suggests adequate fit.
4.5.2 Common method bias (CMB)
Since moderating effects are generally not affected by CMB (Siemsen et al., 2010), this should alleviate concerns that the effects of POS and EPS may be biased. Nevertheless, given that we collected cross-sectional data using a single instrument, we decided that it was necessary to adopt several procedural and statistical remedies to address the issue of CMB (Podsakoff et al., 2012).
For procedural purposes, we made sure that we fully explained the general aim of the study to the participants. We also assured them of the complete anonymity of their responses and informed them that the data collected would only be analysed at an aggregate level (Arnold & Feldman, 1981). In addition, survey items were carefully revised by the CEO of DS, the HR manager, the head of HR development (our key informant), and 10 ESM users. This helped ensure that the questions were clearly formulated and that the respondents would not misinterpret them. It also helped ensure that there was no conceptual overlap between the various items for each construct. Avoiding this type of conceptual overlap is an important way to prevent CMB (Conway & Lance, 2010).
Participation in the study was voluntary, and the respondents were advised that they could end their participation and not finish the survey if they wished. Moreover, to prevent response bias, we presented the survey items in random order. We also used a reversed-item approach to assess the respondents' level of attention while they were answering the survey.
For statistical purposes, we followed the steps recommended by Podsakoff et al. (2003). Harman's single factor test did not reveal any concerns, for there was no single factor that explained most of the variance. Indeed, our results showed that the first-order single factor only explained 19.49% of the total variance, which was significantly lower than the maximum limit of 50% (Podsakoff et al., 2012). In addition, we used the marker variable technique to examine the risk of CMB (Williams et al., 2010). More specifically, we used the construct of perceived fairness, which was not part of our theoretical model, and which could not be expected to have a strong relationship to work-related and nonwork-related knowledge contributions. This robust procedure is well documented in the IS literature (Matook et al., 2015; Pavlou et al., 2007). We examined the correlations between perceived fairness and the other constructs in our model, as this construct could be expected to be impacted by the same common bias as the dependent and the independent variables. The results showed no significant correlations between perceived fairness and the two forms of knowledge contributions (0.087 for work-related knowledge contributions and −0.027 for nonwork-related knowledge contributions). We also created a common latent factor (CLF) and compared the standardised regression weights when we ran the regression with and without this CLF (Mittendorf et al., 2019; Podsakoff et al., 2003). For the purpose of the analysis, we used AMOS 27 IBM SPSS. The difference between the standardised regression weights with and without CLF was less than 0.2. These thorough analyses all indicated that CMB was not a major concern for our results.
4.5.3 Selection bias
To eliminate the risk of endogeneity related to the selection treatment (Hill et al., 2021), we followed the recommendation of Hill et al. (2021), who point out that ‘if the endogenous variable is continuous and selected by the subject or the context, then methods used to address [the issue of a possible] omitted variable are applicable’ (p. 121). The calculation of the impact threshold of a confounding variable (ITCV) was the method that we chose for this purpose. In recent research, this method has been used to eliminate the risk of bias in the selection treatment (O'Toole et al., 2023).
Using this approach also enabled us to verify that there were no omitted variables (Busenbark et al., 2022; Hill et al., 2021). Omitted variables are important variables that have not been included in the analysis even though they can influence the results in a way that changes the relationship between the dependent and independent variables. Busenbark et al. (2022) urge researchers to take measures both before and during the publication process to ensure that bias resulting from omitted variables has not skewed their results. The method based on calculating the ITCV has also been used in IS research (Huang et al., 2017). It consists in calculating the threshold value of correlations at which an omitted variable would falsify research results (Busenbark et al., 2022; Frank, 2000). We used the konfound package in Stata/MP 17.0 to calculate the ITCV. The results of this test yielded an ITCV value of 32.9% at a 10% significance level. This meant that an omitted variable would need to be correlated at higher than 0.329 with both our dependent variable and our independent variable to invalidate our findings. The examination of the correlations between these two variables and our control variables (Table 3) showed that no variable was correlated with the dependent and independent variables at a high enough level for it to have a significant impact as a confounding variable. The strongest correlated variable for work-related knowledge contributions was job field at r = 0.319, which was only correlated at r = 0.205 for nonwork-related knowledge contributions. We also controlled for the ITCV value by examining the partial correlations of the control variables with the dependent and independent variables. Once again, the partial correlations did not exceed the ITCV value. The highest partial correlation with work-related knowledge contributions was r = 0.279, and the highest partial correlation with nonwork-related contribution was r = 0.173, both of which were observed for job field. These values were lower than the ITCV threshold.
Therefore, we maintain that it is very unlikely that there exists a confounding variable that will significantly affect our results. Moreover, the robustness of the inference to replacement (RIR) was 76.30%. This means that 76.30% of our estimate would have to be biased to skew our results. In a word, the risk of endogeneity due to treatment selection appears to be insignificant (Busenbark et al., 2022; O'Toole et al., 2023).
4.5.4 Hypothesis testing
We used Hayes's PROCESS macro for SPSS (Hayes, 2015, 2017; Hayes & Preacher, 2014) to assess the study hypotheses, a technique that is widely used in IS research (Lee et al., 2018; Tams, Ahuja, et al., 2020). We mean-centered all the variables before we entered them into the model (Aiken et al., 1991). Since our model postulates that POS and EPS both moderate the relationship between the two forms of knowledge contributions, we used the model 2 process with a 95% confidence interval and 5000 bootstrap resamples in SPSS version 27, which enabled us to assess the moderating effect of both of these variables (EPS and POS) at the same time.
The results showed that the effect of the interaction between nonwork-related knowledge contributions and POS and the effect of the interaction between nonwork-related knowledge contributions and EPS explained an additional 2.64% of the variance in work-related knowledge contributions. This was over and above what the main model for the effect of nonwork-related knowledge contributions and for the effect of the control variables explained. In other words, adding these two moderating effects provided a further explanation of the relationship between the two forms of knowledge contributions on ESM.
As shown in Table 7, nonwork-related knowledge contributions are positively and significantly related to work-related knowledge contributions (β = 0.2930; SE = 0.0470; t = 6.2332; p = 0.000), a result that fully supports H1. The interaction between nonwork-related knowledge contributions and POS explains an additional 1.65% of the variance in work-related knowledge contributions. This is over and above what is explained by the main model for the effects, which does take into consideration the moderating effects. These results show that the interaction between POS and nonwork-related knowledge contributions is a significant and positive predictor of work-related knowledge contributions (interaction coefficient = 0.1902; SE = 0.0760; t = 2.5018; CI95 [0.0405, 0.3399], not including zero). Therefore, H2 is also fully supported.
| Constructs | β | SE | t | P | LLCI | ULCI |
|---|---|---|---|---|---|---|
| NWKC | 0.2930 | 0.0470 | 6.2332 | 0.0000 | 0.2004 | 0.3856 |
| POS | 0.0540 | 0.0818 | 0.6601 | 0.5098 | −0.1071 | 0.2152 |
| Interaction effect (POS, NWKC) | 0.1902 | 0.0760 | 2.5018 | 0.0130 | 0.0405 | 0.3399 |
| EPS | 0.1131 | 0.0651 | 1.7375 | 0.0835 | −0.0151 | 0.2413 |
| Interaction effect (EPS, NWKC) | −1754 | 0.0565 | −3.1048 | 0.0021 | −0.2866 | −0.0641 |
| OCB | 0.2054 | 0.0744 | 2.7596 | 0.0062 | 0.0588 | 0.3521 |
| Education | −0.0920 | 0.0439 | −2.0975 | 0.0369 | −0.1783 | −0.0056 |
| Age | 0.0021 | 0.0084 | 0.2501 | 0.8027 | −0.0144 | 0.0186 |
| Organisational tenure | 0.0149 | 0.0226 | 0.6589 | 0.5106 | −0.0297 | 0.0595 |
| Gender | 0.0361 | 0.1011 | 0.3574 | 0.7211 | −0.1630 | 0.2352 |
| Rank | −0.1090 | 0.1171 | −0.9307 | 0.3529 | −0.3397 | 0.1216 |
| Job field | 0.1931 | 0.0443 | 4.3624 | 0.0000 | 0.1059 | 0.2803 |
The interaction between nonwork-related knowledge contributions and EPS explains an additional 2.54% of the variance of work-related knowledge contributions. This is over and above what is explained by the main model for the effects, which does not take into consideration the moderating effects. These results show that the interaction between EPS and nonwork-related knowledge contributions is a significant yet negative predictor of work-related knowledge contributions (interaction coefficient = −0.1754, SE = 0.0565, t = −3.1048, CI95 [−0.2866, −0.0641], not including zero). Therefore, despite the significant interaction between EPS and these knowledge contributions, H3 is not supported.
5 DISCUSSION
As a unique form of digital platform, ESM plays a fundamental role in organisational knowledge strategies (Engelbrecht et al., 2019; Neeley & Leonardi, 2018). However, it can only be beneficial to organisations if their employees use it to contribute work-related knowledge that is of value to their colleagues (Leonardi, 2017; Leonardi & Treem, 2012). This is why prior research has focused on the drivers (Kwahk & Park, 2016; Rode, 2016) and the impacts (Beck et al., 2014; Leonardi, 2014) of knowledge contribution behaviour on ESM.
Using SET (Blau, 1964; Cook, 1977) and insights from the APE model (Gawronski & Bodenhausen, 2006, 2014), we develop a conceptual framework that theorises the association between the work and the nonwork domain on ESM. Focusing on work-related knowledge contributions and nonwork-related knowledge contributions as distinct forms of behaviour, we argue that the relationship between these two forms of behaviour is moderated by the EE relationship that binds employees to their employer. The findings that we obtained using the data compiled from the surveys completed by the knowledge workers at the French e-commerce firm confirm that nonwork-related knowledge contributions on ESM positively influence work-related ones. We maintain that this influence is based on reciprocity norms that shape the behaviour of employees who view the organisational decision to adopt an ESM platform as a positive initiative that makes it possible to engage in a form of play at work (Petelczyc et al., 2018)—communicating about nonwork-related personal interests with colleagues while on the job. Through this decision, the organisation signals to its employees that they can pursue their nonwork-related personal interests at work (Howe et al., 2022). This nurtures a sense of indebtedness among employees (Kurtessis et al., 2017) because they greatly appreciate the favourable communications context that the organisation has provided, increasing the likelihood that they will reciprocate (Cropanzano et al., 2017). In other words, employees who use the platform to make nonwork-related knowledge contributions on ESM feel inclined to compensate the organisation by making work-related knowledge contributions, thereby maintaining a social exchange relationship that is equally advantageous to all the parties involved.
Beyond the initiative to adopt an ESM platform, an organisation makes numerous other decisions and engages in a wide range of different actions that can impact how its employees perceive the quality of the social exchange relationship that binds them to the organisation and shape their behaviour on ESM. These decisions and actions also help determine to what extent employees are committed to engaging in reciprocity behaviour. In this study, we conceptualise this EE relationship as a function of POS and EPS, arguing that both of these perceptual beliefs moderate the relationship between the two forms of knowledge contributions. We maintain that employees who frequently contribute nonwork-related knowledge on ESM are more motivated to spend time and effort contributing work-related knowledge on ESM when they perceive organisational support and when they feel that it is safe to express their ideas and opinions in the work environment. That is, the benefits of nonwork-related knowledge contributions are more likely to encourage employees to make work-related knowledge contributions when POS and EPS are high. However, although our findings confirm the positive moderating effect of POS, they do not confirm the hypothesized effect of EPS.
In fact, although our empirical analysis showed that EPS moderates the relationship between the two forms of knowledge contributions, it revealed that it does so negatively, for the relationship was weaker when employees perceived higher levels of psychological safety in the work environment. To investigate this counterintuitive result, we analysed the EPS construct further and realised that it was positively and significatively related to the dependent variable in our model (see Table 7). Thus EPS—the perception that one can express one's ideas and opinions in the work environment without risk of rejection or retaliation—exhibits effects that are similar to those associated with nonwork-related knowledge contributions, which means that with respect to the dependent variable, the two play competing roles instead of mutually supporting it. Indeed, from a theoretical standpoint, it is clear that employee psychological safety is directly related to extra-role behaviour (Singh et al., 2013) and to proactive actions (Detert & Burris, 2007), and that it can help mitigate perceived hierarchical differences within an organisation (Edmondson, 1999). These advantages are also related to nonwork-related knowledge contributions, for these contributions can also be understood as a form of extra-role behaviour (Demerouti et al., 2015) that facilitates serendipitous encounters (Qureshi et al., 2018), making it easier to find colleagues with mutual interests and complementary expertise, and engage with them in ways that cross hierarchal lines (Oostervink et al., 2016). These insights provide a plausible explanation for the counterintuitive moderating effects of EPS revealed by our empirical analysis.
5.1 Theoretical implications
Our study makes three contributions to the literature on knowledge contributions and ESM. First, although the existing research on ESM distinguishes the work domain from the nonwork domain, it conceptualises the latter as an adapted setting that merely enables IT appropriation activities (DeSanctis and Poole 1994) such as blogging (Lu et al., 2015), posting (Huang et al., 2015), and commenting (Laitinen & Sivunen, 2020). We are unaware of any other study that has examined the nonwork domain on ESM as a fertile ground for knowledge contributions. Indeed, there appears to be no prior research on nonwork-related knowledge contributions on ESM as a distinct form of communicational behaviour at work that revolves around discussing topics that have to do with one's nonwork-related personal interests (Howe et al., 2022). Thus, in introducing the concept of nonwork-related knowledge contributions on ESM, our study extends the previous conceptualization of knowledge contributions in online settings (Ma & Agarwal, 2007).
Second, prior research on the relationship between the work and the nonwork domain on ESM has produced mixed results. One study has found that the nonwork-related use of ESM can cause work disruptions (Gibbs et al., 2015). Another study has found that it can distract employees from their work tasks and reduce employee collaboration and performance (Zhang et al., 2023). Yet other researchers have found that posting on nonwork-related ESM channels can stimulate employee curiosity (Neeley & Leonardi, 2018) and that under some conditions, it can boost employee work performance (Lu et al., 2015). Our findings contribute to this debate by showing that even if nonwork-related knowledge contributions on ESM and work-related ones are semantically unassociated, they are closely connected and even positively associated as concrete communicational behaviours that occur in the work environment.
Third, our study contributes to the literature by showing that as fundamentally important aspects of the EE relationship, POS and EPS are key moderating factors for the relationship between the two forms of knowledge contributions on ESM. This result reveals that work-related knowledge contributions are motivated not only by individual (Rode, 2016) or dyadic (Beck et al., 2014) social exchange considerations, but also by favourable organisational conditions that facilitate social exchanges in the work environment. Although the need for facilitating organisational conditions has been previously discussed in the literature (Venkatesh et al., 2008, 2003), the ones that have been examined are mainly related to the availability of technological resources for removing barriers to the adoption and use of information systems. These facilitating organisational conditions are different from the ones that we examine in our study, which are unique attributes of the work environment that moderate important social aspects of organisational life.
5.2 Practical implications
Building on our theoretical contributions, we can outline some of the important managerial implications of our research. First, although nonwork-related online activities are often seen as wasting precious company time (Syrek et al., 2018) and IT resources (Song et al., 2019), our results suggest that this type of negative assumption should be avoided, for they indicate that nonwork-related knowledge contributions can have positive impacts on the work-related ones that have long been considered a fundamental condition for reaping the organisational benefits of ESM (Beck et al., 2014; Leonardi, 2017). It is important to reiterate that in many organisational contexts some employees choose to withhold knowledge despite the numerous attempts of their employers to encourage them to share it (Ghasemaghaei & Turel, 2021). Thus it should come as no surprise that Fortune 500 companies reportedly lose a combined $31.5 billion a year because of employees who fail to share knowledge effectively (Myers, 2017). Given that autonomous motivation is known to be superior to controlled motivation in knowledge contribution contexts (Reinholt et al., 2011), we believe that in order to foster work-related knowledge contributions as a highly desirable reciprocating behaviour, managers should encourage employees to make nonwork-related knowledge contributions while they are on the job.
Moreover, organisations would do well to create a work environment in which employees feel that their nonwork-related knowledge contributions are encouraged and valued. By offering employees a space where they can communicate about their personal passions and interests with their colleagues (Howe et al., 2022), an employer strengthens the social bonds between employees (Lu et al., 2015) and allows them to develop a sense of togetherness that sets the stage for extra-role behaviour (Cropanzano et al., 2017) such as making work-related knowledge contributions (Tams, Ahuja, et al., 2020; Tams, Dulipovici, et al., 2020). Creating such an environment is challenging, but the enactment of appropriate policies can be the spark that ignites enthusiasm for the necessary organisational changes (Bodhi et al., 2021). In particular, an attempt must be made to formalise the beneficial aspects of nonwork-related knowledge contributions by explicitly informing employees that these empowering contributions are encouraged and valued. When organisations do this, they pave the way for the emergence of a supportive work environment that fosters an enduring cognitive association (Chang & Ko, 2016; Gawronski & Bodenhausen, 2014) between the two forms of knowledge contributions on ESM.
Our results also suggest that organisations would do well to create and maintain a work environment in which employees feel safe and supported. Beyond the often-cited benefits of having empowered employees who feel safe from professional risks and who feel valued and cared for, our study brings to light the amplification effects of these beliefs as they can fuel or diminish employees' desire to reciprocate the benefits received from their organisation. Accordingly, POS and EPS need to be viewed by managers as important aspects of a broader organisational and cultural strategy (Frank et al., 2004). These managers should make sure to implement actions and steps that consistently communicate the values these beliefs underline. They should focus on employees' perceptions of support and safety by putting in place organisational programs that alleviate employees' concerns and that demonstrate care and support (Saks, 2006).
5.3 Limitations of the study and avenues for future research
Future research could extend the results of our study by addressing its limitations. First, the study sample consisted of employees from a single firm with a specific organisational culture that differs from other organisational realities. Since organisational culture has been found to impact knowledge behaviour (Alavi et al., 2005), we recommend that future studies attempt to use our research model to examine work-related and nonwork-related knowledge contributions at firms with different organisational cultures.
Second, most of the DS employees belong to the millennial generation for whom social media is the preferred communications technology (Meghan, 2016). Since organisations often have employees with a much greater range of ages, future research could extend our study by creating a research design to compare the knowledge-sharing behaviour of employees who are millennials with the knowledge-sharing behaviour of employees from other generations. Such a comparison is worth exploring because recent research has shown that cognitive age has an important impact on the use of IT by employees (Tams, 2022).
Third, our study does not take into consideration the intensity of work even though it can help determine the likelihood that employees will make either work-related or nonwork-related knowledge contributions on ESM. To address this limitation, future research could use role-based manipulation that randomly assigns employees to a high-intensity or to a low-intensity work context before they are asked questions about contributing work-related knowledge and nonwork-related knowledge on ESM.
Fourth, we used self-report items to measure employees' overall assessment of their knowledge contribution behaviour, but we did not take into consideration the evolution of this behaviour over time (Ghasemaghaei & Turel, 2021; Qureshi et al., 2018). Since the knowledge contribution behaviour of employees can evolve as they progress in the organisational hierarchy, future research could examine the extent to which employee promotions and career changes impact work-related and nonwork-related knowledge contributions on ESM.
Fifth, given the emerging interest in the impact of social media on employee well-being (Bodhi, Singh, & Joshi, 2022; Luqman et al., 2021), our findings could be extended by investigating how and when nonwork-related knowledge contributions impact employee well-being.
Finally, at the firm that provided the research context for our study, the only ESM platform used to host the two forms of knowledge contributions was Slack. Yet some organisations offer their employees the opportunity to use multiple ESM platforms (Forsgren & Byström, 2017), even encouraging them to use their private social media accounts for work-related purposes (Bodhi, Luqman, et al., 2022) and to use both ESM and their private social media accounts for nonwork-related purposes (Liang et al., 2020). Therefore, future research could examine the impact of the two forms of knowledge contributions on key employee outcomes in the context of various social media configurations.
6 CONCLUSION
In the new world of remote work (Tarafdar & Saunders, 2022), where work-from-anywhere arrangements (Raj Choudhury et al., 2021) have become the norm, the use—and the usefulness—of ESM can be expected to continue increasing due to the capacity of this type of platform to support both work-related and nonwork-related knowledge contributions. We hope that our study's findings have provided theoretical and practical insights that will pave the way for a deeper examination of the ever-increasing integration of the nonwork domain into the work domain, one that will be beneficial to employers and employees alike as they grapple with the complexities of this integration.
ACKNOWLEDGEMENTS
We are grateful for the guidance of the Senior Editor, the Associate Editor, and all the reviewers for their constructive feedback and developmental reviews during the rounds of revisions. We also thank Stefan Tams for his valuable advice and suggestions on prior versions of the article.
APPENDIX A: Literature review: Enterprise social media and knowledge contributions.
| Authors | Distinction between the two forms of Knowledge contributions on ESM | Role of the organisation in the relationship between the work and the nonwork domains on ESM | Conceptual framework or theoretical background | Methodology and research context | Variables and relationships | Main findings |
|---|---|---|---|---|---|---|
| Ren and Sun (2023) | No distinction is made between the two forms of knowledge contributions. | The role of the organisation is not examined. | Social Cognitive Theory | Survey of users of ESM conducted in China with the assistance of the Alumni Office of the authors' university | IV: Psychological Safety and Psychological Empowerment Mediating Variable: ESM Affordances Moderating Variable: Project Task Context DV: Effectiveness of Knowledge Sharing |
ESM affordances have a positive effect on the effectiveness of knowledge sharing. Psychological safety and psychological empowerment of team members have a significant positive influence on ESM affordances. Project task complexity positively moderates the positive effects that the affordances of visibility and association have on the effectiveness of knowledge sharing and negatively moderates the positive relationship between the affordance of editability and the effectiveness of knowledge sharing |
| Dwivedi et al. (2022) | No distinction is made between the two forms of knowledge contributions. | The role of the organisation is not examined. | Dynamic Capabilities, the Resource-based View, and the Knowledge-based View | Survey of public- and private-sector organisations in Bangladesh that used ESM during the COVID-19 virus outbreak for work-related reasons. | IV: ESM Usage Mediating Variable: Knowledge Sharing via ESM DV: Operational and Social Performance |
Organisational agility and ESM infrastructure are strong mediators of the link between ESM usage and decision-making effectiveness, while the mediating effect of knowledge sharing through ESM is weaker. |
| Laitinen and Sivunen (2020) | No distinction is made between the two forms of knowledge contributions. | The role of the organisation is examined in terms of norms, tasks, and repertoires. | Theories of Communication Privacy Management Theory and Technology Affordance | Qualitative field study based on semi-structured interviews and enterprise social media review data from a large Nordic media organisation. | NA | Three categories of antecedents influence employee decision to share information on ESM: personal dimension (personal privacy boundaries, professional boundaries and risks, online safety, and perceived audience), technological dimension (ex: visibility, persistence, awareness and searchability), and organisational dimension (norms, tasks, and repertoires) |
| Chin et al. (2020) | No distinction is made between the two forms of knowledge contributions. | The role of the organisation is not examined. | The Unified Theory of Acceptance and Use of Technology (UTAUT) | Survey data was collected from 158 employees whose organisations are currently deploying an ESM | IV: Content Value, Relationship Expectancy and the UTAUT's Original Constructs DV: Contribute ESM Use, Consumptive ESM use, Usage Gap, and Overall ESN Use |
The contributive use on ESM is strongly impacted by the antecedents of social influence, content value and relationship expectancy. |
| Sun et al. (2020) | No distinction is made between the two forms of knowledge contributions. | The role of the organisation is not examined. | Affordance Theory | Empirical data was collected via survey responses gathered from 365 Chinese employees who uses Dingtalk | IV: Affordances: Association, Visibility, Persistence and Editability Mediating Variable: Knowledge Provision and Knowledge Acquisition DV: Creative Performance |
All ESM affordances (association, visibility, persistence, and editability) were found to have a significant and positive impact on employee knowledge provision on ESM. |
| Sun et al. (2019) | No distinction is made between the two forms of knowledge contributions. | The role of the organisation is not examined. | Affordance Theory | Conceptual paper that examines integrates the association between ESM affordances and knowledge sharing | N/A | Prior work considered artefacts without allocating much attention to the roles of individual goals and organisational context. ESM can both enable and hinder knowledge sharing by affording different user behaviours that depends on artefacts, individual goals and organisational context. |
| Engelbrecht et al. (2019) | No distinction is made between the two forms of knowledge contributions. | The role of the organisation is not examined. | Communication visibility theory | Quantitative method with a focus on a panel of knowledge workers who use ESM at work in Germany and Switzerland. | IV: ESM use Mediating Variable: Communications awareness Moderating Variable: Management Responsibility DV: Metaknowledge |
The creation of meta-knowledge through ESM occurs across contexts. ESM leads to an immediate increase in meta-knowledge, with managers gaining more meta-knowledge than regular employees. |
| Neeley and Leonardi (2018) | The work and nonwork domains on ESM are clearly identified, but there is no explicit distinction between the two forms of knowledge contributions. | The role of the organisation in enabling a transition between nonwork interactions and work-related knowledge contributions is qualitatively explored. | Strategy-as-practice perspective | Two qualitative case studies conducted at two large American organisations. | N/A | Organisations can play a negative and also a positive role in shaping the transition between the nonwork domain and the work domain on ESM. |
| Kane (2017) | No distinction is made between the two forms of knowledge contributions. | The role of the organisation is not explicitly examined, but the case of a European cosmetic company whose bureaucratic culture had a negative impact on collaboration and knowledge-sharing on ESM is discussed. | Knowledge management | Conceptual article | N/A |
Social media is not a uniform set of technologies but instead a diverse array of continuously evolving technological systems for human communication and collaboration. |
| Leonardi (2017) | No distinction is made between the two forms of knowledge contributions, but it is argued that nonwork-related interactions on ESM can seed work-related ones. | The role of the organisation is not examined. | IT affordances | Two qualitative case studies conducted at two large American organisations | N/A |
Encouraging the sharing of nonwork-related content on ESM can seed work-related interactions on ESM. |
| Rode (2016) | No distinction is made between the two forms of knowledge contributions. | The role of the organisation is not examined. | Social exchange theory | Survey of employees who use ESM at a large German multinational high-tech firm in the B2B sector | IV: Reputation, reciprocal benefits, enjoyment in helping others, and self-efficacy in knowledge sharing DV: Knowledge sharing on ESM |
Anticipated gains in reputation, reciprocal benefits, and knowledge-sharing self-efficacy lead employees to share knowledge on ESM. |
| Oostervink et al. (2016) | No distinction is made between the two forms of knowledge contributions. | The role of the organisation is not examined. | Institutional theory | Qualitative case study of an international IT consulting firm present in 50 unnamed countries | N/A |
When using ESM, consultants are confronted by knowledge-sharing ambiguities because of institutional complexity related to an opposition between a professional logic and a corporative logic. |
| Mäntymäki and Riemer (2016) | A distinction is made between the social and the hedonic aspects of ESM use, but no distinction is made between the two forms of knowledge contributions. | The role of the organisation is not examined. | Cognitive dissonance theory and public goods theory | Qualitative data on ESM use collected at three Australian firms, and quantitative data on ESM use collected at five Australian firms. In all cases, ESM use is part of day-to-day work practices. | IV: Discussion of ideas and work issues, problem solving, task management, events and updates, and informal talk DV: Value of ESM use. |
Employees value ESM for knowledge sharing and informal discussions on ESM, and this is positively associated with the utilitarian use of ESM. |
| Leonardi (2015) | No distinction is made between the two forms of knowledge contributions. | The role of the organisation is not examined. | Affordances | Quasi-natural field experiment conducted at a large financial services firm headquartered in the American Midwest | IV: Exposure to content of others' messages and exposure to indicators of others' communication partners, tenure, hierarchical level, number of close colleagues, number of team members, and advice network centrality DV: Accuracy of who knows what and accuracy of who knows whom |
With ESM, employees can know about the communications between their colleagues, and this makes it easier for them to make inferences about what and whom these co-workers know. |
| Ellison et al. (2015) | No distinction is made between the two forms of knowledge contributions. | The role of the organisation is not examined. | Affordances | Conceptual article | N/A |
ESM affordances can shape knowledge sharing by bringing into play social capital dynamics, by supporting relationships and interactions, and by helping users navigate context collapse. |
| Leonardi (2014) | No distinction is made between the two forms of knowledge contributions. | The role of the organisation is not examined. | Communication visibility | Thirty-four semi-structured interviews with marketing and operations management employees at a large financial services firm headquartered in the American Midwest. | N/A |
Enhanced meta-knowledge can be used to help create more innovative products and services, and to help reduce knowledge duplication while employees are learning new work practices. |
| Beck et al. (2014) | No distinction is made between the two forms of knowledge contributions. | The role of the organisation is not examined. | Social exchange theory, social identity theory, and theory of power reduction | Analysis of the messages posted on an ESM platform at a leading multinational service provider in an unnamed country | IV: Reputation, habit of cooperation, group identification, social status, variety of channels, social presence, reciprocity norms, indebtedness, relationship strength, intellectual closeness, and cultural closeness DV: Quality of the knowledge exchanged |
On ESM, the sociocultural characteristics of knowledge seekers as well as relational factors determine the quality of knowledge exchanges. |
| Gibbs et al. (2013) | No distinction is made between the two forms of knowledge contributions. | The role of the organisation is not examined. | Dialectical theory | Qualitative case study of the engineering division at a distributed high-tech startup in the United States | N/A | On ESM, employees share knowledge while at the same time managing tensions between visibility and invisibility, between engagement and disengagement, and between imparting information and maintaining control over it. |
APPENDIX B: Measurements items.
| Construct | Items | Adaptations and sources |
|---|---|---|
Work-related knowledge contributions (WKC) |
WKC1: I often help colleagues who ask for assistance or information from other members of the organisation in work-related channels on Slack. WKC2: I actively participate in work-related channels on Slack. WKC3: I have shared knowledge in work-related channels on Slack. WKC4: I have shared knowledge in work-related channels on Slack with other members of the organisation, and this has allowed them to develop new insights. |
These items are adapted from Ma and Agarwal (2007) study of online communities to the fit an ESM work context. |
| Nonwork-related knowledge contributions (NWKC) | NWKC1: I often help colleagues who need assistance or information from other members of the organisation in nonwork-related channels on Slack. NWKC2: I actively participate in nonwork-related channels on Slack. NWKC3: I have shared knowledge in nonwork-related channels on Slack. NWKC4: I have shared knowledge in nonwork-related channels on Slack with other members of the organisation, and this has allowed them to develop new insights. |
These items are adapted from Ma and Agarwal (2007) study of online communities to fit the context of a nonwork-related online community on ESM. |
| Perceived organisational support (POS) | POS1: Digital Success values my contribution to organisational well-being. POS2: Digital Success makes a strong effort to take into consideration my goals and values. POS3: Digital Success really cares about my well-being. POS4: Digital Success is willing to help me when I need a special favour. POS5: Digital Success shows very little concern for me (reverse coded). POS6: Digital Success takes pride in my accomplishments at work. |
These items are adapted from the measures used by Rhoades et al. (2001). In particular, the word ‘organisation’ has been replaced by the name of the firm: ‘Digital Success’. |
| Employee psychological safety (EPS) | EPS1: It is safe to give my opinions at work. EPS2: It is safe for me to speak up around here. EPS3: It is safe for me to make suggestions at work. |
These items are taken from Burris et al. (2008). |
| Organisational citizenship behaviour (OCB) toward colleagues | OCB1. I help collaborators who have been absent. OCB2. I help collaborators who have a heavy workload. OCB3. I help integrate new people even though I am not required to do so. OCB4. I willingly help collaborators who have work-related problems. OCB5. I am always ready to lend a helping hand to those around me. (This item was dropped.) |
These items are adapted from the measures used by Ong et al. (2018). In particular, the word ‘others’ has been replaced by the word ‘collaborators’. |
Biographies
Nabila Boukef is an Associate Professor of Information Systems and Digital Transformation at SKEMA Business School, Université Côte d'Azur, France. She received her Ph.D. from Paris IX Dauphine University, France. Her research interests include media use in the workplace and collaboration in dispersed teams. She received the best Ph.D. Award in Information Systems in France and the best reviewer as well as the best paper awards in Systèmes d'Information & Management. Her work appeared in different outlets including the Journal of Strategic Information Systems, Journal of Business Ethics, and Systèmes d'Information & Management. Academy of Management Annual Meeting (AOM), International Conference on Information Systems, and European Conference on Information Systems.
Mohamed Hédi Charki is a Professor of Information Systems at EDHEC Business School. He holds a Ph.D. in Management Science from Paris IX Dauphine University. His research interests revolve around social networks, employee creativity and employee well-being. He has published in outlets such as the Journal of Management Information Systems, Journal of Strategic Information Systems, and Systèmes d'Information & Management. Mohamed Hédi Charki obtained the Best Associate Editor Award from the International Conference on Information Systems (Social Media & Digital Collaboration track) in 2020 and earned the Best Reviewer Award from the Academy of Management Conference (Communication, Digital Technology & Organisations division) in 2023. He serves as an Associate Editor of Information & Management.
Mustapha Cheikh-Ammar is an Associate Professor of Information Systems at the Faculty of Business Administration, Laval University. Before joining Laval, he was an Assistant Professor at Ivey Business School. He earned his PhD in Business Administration from HEC Montreal, where he was also a member of the Canada Research Chair in Information Technology Implementation and Management. His research has been published or accepted for publication in European Journal of Information Systems, IEEE Software, Information & Management, International Journal of Consumer Studies, Journal of Organisational and End User Computing, Knowledge Management Research and Practice, and Technological Forecasting and Social Change among other journals and conference proceedings.
Open Research
DATA AVAILABILITY STATEMENT
We cannot share the data because of the confidentiality agreement with the company.
