Work‐related social media use and the shaping of communicative role perceptions

This is a self-archived version of an original article. This version may differ from the original in pagination and typographic details. Author(s): Title: Year: Version: Copyright: Rights: Rights url: Please cite the original version: CC BY 4.0 https://creativecommons.org/licenses/by/4.0/ Work‐related social media use and the shaping of communicative role perceptions © 2023 the Authors Published version Pekkala, Kaisa; van Zoonen, Ward Pekkala, K., & van Zoonen, W. (2023). Work‐related social media use and the shaping of communicative role perceptions. European Management Review, Early View. https://doi.org/10.1111/emre.12597 2023 RESEARCH ARTICLE Work-related social media use and the shaping of communicative role perceptions Kaisa Pekkala 1 | Ward van Zoonen 2 1 Department of Social Sciences, LUT University, Lappeenranta, Finland 2 Jyväskylä University School of Business and Economics (JSBE), Jyväskylä, Finland Correspondence Kaisa Pekkala, Department of Social Sciences, LUT University, Lappeenranta, Finland. Email: [email protected] Abstract This study focuses on employees’work-related social media use. The multivalent involvement of social media in corporate processes calls for attention to how employees’social media use is conceptualized and managed. Drawing on a sample of 1179 knowledge workers, the study explores how employees perceive their communicative roles, how contextual factors shape these perceptions, and how communicative role perceptions, in turn, are associated with work-related social media use. The findings demonstrate that leadership support and employees’perceptions of the anticipated impact of their communication are positively related to role perceptions, and the more employees define their communicative roles as an expected part of their work, the more likely they are to use social media for work-related purposes. This study is highly relevant to scholars and managers as it draws attention to expanding workplace roles that emerge in relation to advancements in, and the adoption of, new information and communication technology. KEYWORDS contextual behavior, identification, role perceptions, social media INTRODUCTION In recent years, one of the most significant changes in the workplace has been the evolution and increased use of communication technologies (Colbert et al., 2016; Leonardi & Vaast, 2017). Social media have had a particularly profound impact on the nature of knowledge work by allowing anyone to create and share work-related information to broad audiences within and across organizational boundaries (Leonardi & Vaast, 2017). The opportunities provided by social media have generated new expectations about employees’use of these technologies (Andersson, 2019b; Pekkala et al., 2022; Pekkala & van Zoonen, 2022; Walden, 2018), resulting in the emergence of new communicative roles for employees (Madsen & Verhoeven, 2019). Employees’online behavior has become an integral part of the organization’s outward representation, making individual employees “ambassadors for their organization and its cause”(Siegert & Löwstedt, 2019,p.33).Most studies on employees’work-related social media communication seek to explain this type of behavior as organizational citizenship behavior (OCB) and discretionary in nature, subject mostly to individual employees’ needs or intentions (e.g., Boukis & Christodoulides, 2020; Fieseler et al., 2015; Helm et al., 2016;Men,2014). However, recent studies indicate that not all employees voluntarily choose to display these behaviors; instead, they consider an ambassadorship role as a work-related responsibility (Andersson, 2019a,2019b). Employees’work-related social media use has become a critical success factor for organizations, especially in the knowledge sector, and it has been linked to assets such as brand equity (Boukis & Christodoulides, 2020)andsocial capital (Nikolopoulos & Dana, 2017). According to Pew Research (2016), most employees (77%) in the United States report using social media at work. Moreover, the time spent using social media has increased worldwide, averaging 151 min of social media use daily in 2022 (Statista, 2023). Given the global magnitude of the phenomenon, there is an urgent need for empirical research on the mechanisms that shape employees’ DOI: 10.1111/emre.12597 This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. © 2023 The Authors. European Management Review published by John Wiley & Sons Ltd on behalf of European Academy of Management (EURAM). European Management Review. 2023;1–15. wileyonlinelibrary.com/journal/emre 1 communicative role perceptions and, consequently, their online communicative behavior. Role perception refers to a cognitive process whereby employees “define their work role, including what types and breadth of tasks, goals, and problems they see as within their set of responsibilities, and how they believe they should approach those tasks, goals, and problems to be effective”(Parker, 2007, p. 404). This study focuses on the mediating role of employees’ communicative role perceptions and provides a novel perspective on how social media, as an emerging context for work-related communication and interaction, may shape work-related roles and behavior. We draw on role theory (Biddle, 1979; Morrison, 1994) and literature on work-related social media use (Dreher, 2014; Oksa et al., 2021; Pekkala & van Zoonen, 2022;Van Zoonen et al., 2016) to identify antecedents of employees’ communicative role perceptions and work-related social media use. We identify three factors (i.e., leadership support related to one’s use of social media for professional purposes, perceived impact of communication, and organizational identification indicating one’s willingness to represent the organization) that have been linked to workrelated social media use and employees’role perceptions. Role theory suggests that employees’behaviors and attitudes are shaped by the context in which they operate, including formal job duties and expectations and norms associated with the work role (Morrison, 1994). First, leadership support can provide resources and guidance that empower employees to take on additional roles and expand role perceptions (Andersson, 2019a; Pekkala & van Zoonen, 2022). Research has indicated that organizational leaders increasingly support employees’work-related social media use (Dreher, 2014; Pekkala, 2020) as such communication may contribute to attaining organizational goals (Schaarschmidt & Könsgen, 2020; Siegert & Löwstedt, 2019). By implementing supporting practices and processes, managers gradually increase the expectations of related behaviors, enhancing employees’communicative role (Andersson, 2019b). Second, employees have been found to be aware of the potential impact of their online communication (Helm, 2011). We refer to the perceived impact of communication to denote employees’perceptual awareness of the anticipated impact of their online communication on their employer’s reputation (Helm, 2011). As employees’social media use is partly informed by goal orientations and rational decisions about the efficiency of such use (Leftheriotis & Giannakos, 2014), the perceived impact of communication can influence role perceptions by highlighting the value of effective online communication for employees (Schaarschmidt & Walsh, 2020). Third, organizational identification has been found to predict employees’work-related social media use (Van Zoonen & Banghart, 2018; van Zoonen & Treem, 2019). Organizational identification refers to the degree to which employees define themselves as a member of an organization (Mael & Ashforth, 1992). Identification has also been found to contribute to employees defining their jobs in a broad manner (Morrison, 1994), suggesting that employees who identify with their organizations are typically willing to “do extra”to contribute to their organization’s goals. Engaging in social media use for work can be a way to contribute to such goals (van Zoonen & Treem, 2019). We integrate these perspectives using role theory, suggesting that leadership support, the perceived impact of communication, and organizational identification influence employees’role perceptions by creating an environment that is conducive to taking on new communicative roles. We argue that employees’communicative role perceptions are central to understanding work-related communication behavior. Hence, we propose a novel approach to understanding how different contextual conditions may shape the communicative role perceptions that may underlie work-related social media use (see Figure 1for a conceptual model). The findings of this study will duly contribute to a better understanding of the antecedents and consequences of employees’ changing role perceptions in the digital era. THEORY Role perceptions Role theory suggests that everyday human activities, including activities in the workplace, are guided by socially defined roles (Merton, 1957). According to role theory, expectations define roles within a particular social structure (Biddle, 1979;Katz,1964;Katz& Kahn, 1978). To this end, roles are closely connected to expectations (Ilgen & Hollenbeck, 1991;Katz& Kahn, 1978), awareness of those expectations in specific contexts (Biddle, 1979), and identity (Haslam & Ellemers, 2005). Hence, roles are beliefs that are constituted in, and influenced by, context and social interaction (Parker, 2007;Stryker,2002). Although the idea of role definitions for guiding behavior has “long been advocated and is theoretically plausible,”the relationship between employees’role perceptions and their performance has received relatively little attention (Parker, 2007, pp. 408–409). In the work domain, roles have typically been divided into two principal categories, in-role behaviors, and extra-role behaviors. In-role behaviors are related to task-specific roles and expected behaviors that directly contribute to the core offering of the organization and are typically embedded in formal job requirements (Van Scotter, 2000). Extra-role behaviors relate to “individual behavior that is discretionary, not directly or explicitly recognized by the formal reward system, and that in aggregate promotes the effective functioning of the organization”(Organ, 1988, p. 4). 2PEKKALA and VAN ZOONEN 17404762, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/emre.12597 by University Of Jyväskylä Library, Wiley Online Library on [18/08/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License Employees’role perceptions refer to how an individual construes their work role and how broadly they perceive their role—“what types of tasks, goals, and problems they see as relevant to their role”(Parker, 2007,p.406).Perceived job breadth is a dimension of role perception (Morrison, 1994;Parker,2007), referring to the extent to which employees themselves perceive their behaviors as either an expected part of their work (in-role behavior), or as discretionary behaviors that exceed job requirements (extra-role behavior) (Morrison, 1994). Research has shown that employees are more likely to perform behaviors when they view them as an obligation, rather than as discretionary (Coyle-Shapiro et al., 2004;Morrison,1994), as the former are more likely to be linked to extrinsic rewards and sanctions, both formal and informal (Katz, 1964;Organ,1988). We argue that conceptualizing work-related communicative behaviors as OCB on digital platforms is problematic, or at least too narrowly focused. The entanglement of communication technologies in everyday work practices has expanded the communicative space of work and the rhetorical roles of workers (Huang et al., 2013). This has led to an expansion of role expectations, especially within the knowledge sector (Dekas et al., 2013; Nahapiet et al., 2005). The role literature has long recognized that various factors shape role perceptions and role behaviors (Biddle, 1979). Specifically, task context and social context have been found to be important in shaping role perceptions (Johns, 2006). The task context encompasses the structural and informational opportunities and constraints under which work roles are performed. For example, research found that when employees perceive that communication is critical for organizational success, they tend to take more communication responsibility (Andersson, 2019a). The social context encompasses the relationships and social contingencies, such as social support, that affect role performance (Johns, 2006). For example, according to the role literature, individuals who perceive that their supervisors value and support certain behaviors are more likely to define these behaviors as an expected part of their work (Biddle, 1979; Morrison, 1994). Additionally, when individuals identify with the organization more strongly, they are found to define their jobs more broadly (Morrison, 1994). Hence, drawing on the role literature, we suggest that task context (here: the perceived impact of communication) and social context (here: leadership support and organizational identification) are important in shaping an individual’s communicative role perceptions, and, ultimately, their work-related social media use. DEVELOPMENT OF HYPOTHESES Role perception and its effect on behavior The changing nature of knowledge work, due in part to digital technologies (Dekas et al., 2013), is accompanied by novel expectations about how these technologies shape role requirements and facilitate role fulfillment. Following role theory, we assume that employees differ in how they define their communicative roles and responsibilities (Morrison, 1994; Parker, 2007). Differences in role perception occur because employees’interpretations of role breadth vary. Role breadth refers to whether one regards particular behaviors as an expected—and required—part of one’s work or not (Morrison, 1994; Parker, 2007). Employees’work-related social media communication has been described as (online) ambassadorship behaviors (Andersson, 2019a; van Zoonen et al., 2018), employee advocacy (Men, 2014), brand-building behaviors (Löhndorf & FIGURE 1 Conceptual model. Note: H2, H3, and H4 represent indirect effects of leadership commitment and support (H2), organizational identification (H3), and perceived impact of communication (H4) on work-related social media use through communicative role perceptions. EMPLOYEES’COMMUNICATIVE ROLE PERCEPTION 3 17404762, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/emre.12597 by University Of Jyväskylä Library, Wiley Online Library on [18/08/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License Diamantopoulos, 2014), brand citizenship behavior (Chiang et al., 2012;Helmetal.,2016), and supportive behavior in social media (Schaarschmidt & Könsgen, 2020). These workrelated communication behaviors range from producing, sharing, and following work- and organization-related content through publicly available platforms (e.g., LinkedIn, Twitter, or Facebook) to the creation and maintenance of useful professional networks (Dreher, 2014;Oksa et al., 2021;Pekkala,2020; Van Zoonen et al., 2016). Accordingly, social media are increasingly important in organizational communication, likely shifting perceptions about the nature of such behaviors in the context of job roles. This is important as the role perception literature suggests that when employees define behaviors as an expected part of their jobs, they are more likely to enact these behaviors than if they were defined as discretionary (Coyle-Shapiro et al., 2004; Morrison, 1994). Hence, we hypothesize that broader communicative role perceptions are positively associated with work-related social media use. H1. The more broadly employees define their communicative role, the more likely they are to engage in work-related social media use. Leadership support, role perceptions, and social media use Role theory holds that social context predicts how employees perceive their roles (Johns, 2006) and that employee behavior is influenced by their interpretation of the role expectations they receive from influential others (e.g., direct supervisors) (Katz & Kahn, 1978). The role literature asserts that organizational leaders influence individuals’perceptions and cognitions with respect to their roles and responsibilities, thereby legitimizing and regulating specific behaviors (Biddle, 1979; Graen, 1976). It has also been found that leadership commitment signals the usefulness of technology at work and thus shapes role expectations regarding the use of technologies (Lewis et al., 2003). Several studies provide theoretical and empirical corroboration for the assumption that leadership support may expand employees’role perceptions, including communicative roles. For instance, research demonstrated that employees are likely to reciprocate supervisory support by engaging in behaviors that promote individual and organizational functioning (Eisenberger et al., 2001). This is particularly true in the realm of communicative behaviors such as organizational silence (Morrison & Milliken, 2000), promotive and prohibitive voice behaviors (Liang et al., 2012), and new voice behaviors—namely, social media use (Holland et al., 2016). Research showed that support received from one’ssupervisorisconnectedto employees’predisposition toward taking communication responsibility (Andersson, 2019a). Duan et al. (2017) demonstrated that leaders’voice expectations were positivelyrelatedtoemployees’voice role perceptions. We propose that leadership support positively relates to employees’ felt role obligations (Eisenberger et al., 2001) and communication responsibility (Andersson, 2019a). In turn, following the principles of role theory, when communicative responsibilities are perceived as a more integral part of the role, employees will be more likely to utilize social media for work. For instance, Hansen and Levin (2016) proposed that employees’social media use on behalf of their organization is dependent upon their job roles. Hence, we suggest that leadership support is related to work-related social media use through employees’communicative role perceptions. H2. Leadership support is positively related to work-related social media use through communicative role perceptions. Organizational identification, role perceptions, and social media use The role perception literature suggests that employees who are more attached to their organizations define their jobs more broadly (Morrison, 1994) because of their “desire to protect, support, and improve the organization that surfaces when organizational identities and selfidentities converge”(Hughes & Ahearne, 2010, p. 85). Organizational identification can thus motivate members to act on behalf of the group’s best interests and, as such, identification has been linked to a broader role definition (Riketta, 2005). Moreover, organizational identification has been considered one of the key determinants of employees’advocacy behaviors on social media (e.g., van Zoonen et al., 2018; van Zoonen & Treem, 2019). Research has demonstrated that when individuals identify with their organization, they are more likely to appear online with their professional identity, and also that individuals who perceive a stronger bond with their organization feel more confident in using social media (Fieseler et al., 2015). In addition, it has been shown that when an employee feels their workplace is perceived positively by society, they are more likely to share organizational affiliation across social media (Piercy & Carr, 2023), indicating that online identities are created within a specific context (Fieseler et al., 2015). Along these lines, we predict that employees who identify more strongly with their organizations define their communicative roles more broadly and, subsequently, are more active social media communicators. Hence, we hypothesize that: H3. Organizational identification is positively related to work-related social media use through communicative role perceptions. 4PEKKALA and VAN ZOONEN 17404762, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/emre.12597 by University Of Jyväskylä Library, Wiley Online Library on [18/08/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License Perceived impact of communication, role perceptions, and social media use According to role theory, the task context affects role perceptions and role behavior (Johns, 2006;Mowday& Sutton, 1993). This requires that individuals are aware of the consequences of their respective behavior (Biddle & Thomas, 1966; Katz & Kahn, 1978) and the function of their roles within the social system (Biddle, 1979). Research has found that when employees perceive that communication is critical for organizational success, they tend to take communication responsibility (Andersson, 2019a). This suggests that when employees perceive that their communication can make an impact by contributing to organizational goals, they will be more likely to view such communicative behaviors as part of their job roles. Research has demonstrated that employees who are aware of the potential impact of their communication tend to use social media to represent their organization (Helm, 2011; Schaarschmidt & Walsh, 2020; Siegert & Löwstedt, 2019). In addition, employees’social media use has been informed by goal orientations and rational decisions about efficiency (Leftheriotis & Giannakos, 2014). Furthermore, research has previously linked employees’ social media use to individual role perceptions (Schaarschmidt & Könsgen, 2020). Following role theory, if employees expect that communicating on social media will help them achieve important individual or organizational goals, they may view social media as a part of their job responsibilities (Biddle, 1979). As such, we hypothesize that the perceived impact of communication may lead workers to expand their role perceptions to include such behaviors and therefore drive the use of these technologies to achieve work-related goals. Hence, we hypothesize that H4. The perceived impact of communication is positively related to work-related social media use through communicative role perceptions. METHOD Sample and procedure We tested our hypotheses (see Figure 1) by conducting a survey among knowledge workers in three professional service organizations operating in Finland in the areas of management consultancy, banking, and insurance. The decision to focus on these areas was based on earlier literature suggesting that organizations operating in the knowledge sector are inherently dependent on their employees’communicative agency and ability to communicate with external stakeholders, such as customers and the general public (Alvesson, 2004; Treem, 2016). As a leader in the EU’s Digital Economy and Society Index (DESI), Finland provides an interesting context for studying role perceptions related to social media communication. Finnish companies are active users of social media (51% against the EU average of 29%) (DESI, 2022) and, based on previous studies, Finnish employees use social media for knowledge sharing, information retrieval, and networking and collaboration (Lyly-Yrjänäinen, 2018). All employees of the respective organizations were considered eligible to participate in the study, and an invitation to take part in the online survey was sent to them (n=9786) through email and internal communication channels. All completed surveys were retained for further analysis. A total of 1179 employees completed the survey between November 20 and January 12, 2020. The employees were distributed as follows: 817 worked in the insurance field (company A), 211 in the field of management consultancy (company B), and 151 in the banking field (company C). The response rates in company A and company B were similar at 22% and 17%, respectively. Notably, the response rate in company C was relatively low at 3%; this may have been due to the timing of the survey just before and during the holiday season. The overall response rate across the three companies was 12%. Sixty-one percent of the respondents were female, and 52% were between 30 and 49 years old. Gender distribution did not differ across the companies (χ 2 [4] =5.75, p=219). In company B, the youngest age group was more strongly represented in the data compared with company A and company C—namely, company A 16%, company B 43.6%, and company C 17.2%—χ 2 (8) =104.35, p< .001. In our sample, 38% had been working for their current organization for 1–5 years, 16% for between 6 and 10 years, and 32% for over 10 years. Comparing these percentages across the three case companies did not demonstrate significant differences (χ 2 [2] =3.12, p=211). Ninety-one percent of the respondents had a permanent employment contract, which did not differ across the three companies (χ 2 [2] =1.43, p=488). Measures Leadership support refers to the extent to which employees feel that their leaders are committed to employees’work-related social media communication and encourage and support their communication on social media. Five items measuring perceived leadership commitment and support were derived from Lewis et al. (2003) and assessed with a 5-point Likert scale, ranging from 1 (strongly disagree) to 5 (strongly agree). This construct included items such as “My manager encourages the use of social media at work.” Perceived impact of communication measures employees’ perception of the anticipated impact of their online communication on their employer’sreputation(Helm,2011). The perceived impact was measured using a sub-dimension of EMPLOYEES’COMMUNICATIVE ROLE PERCEPTION 5 17404762, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/emre.12597 by University Of Jyväskylä Library, Wiley Online Library on [18/08/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License the empowerment measure validated and operationalized by Spreitzer (1995). The scale included three items such as “My communicative actions have a significant influence on how my organization is perceived as an employer”and was assessed with a 5-point Likert scale, ranging from 1 (strongly disagree) to 5 (strongly agree). Organizational identification refers to the perception of oneness with or belongingness to an organization (Ashforth & Mael, 1989), an employed measure derived from Smidts et al. (2001) (e.g., used by Fieseler et al., 2015). Five statements including “I experience a strong sense of belonging to my organization”were used to measure organizational identification, and the responses were anchored from 1 (strongly disagree) to 5 (strongly agree). Communicative role perception refers to the extent to which an employee perceives that work-related communication in social media is an expected component of their job. Following the procedure described by Morrison (1994), respondents were asked to assess whether they felt that a particular activity described in the statements was an expected part of their job. Respondents were instructed as follows: We are not interested in whether you perform these activities. Rather, we are interested in whether you yourself see them as part of your job. The activities were specified based on the typology of work-related social media use (Van Zoonen et al., 2016). Contrary to Morrison (1994), responses were inferred on a continuous answer scale (rather than dichotomous). As social media is relatively novel territory in many organizations (Pekkala, 2020), employees’perceptions about social media use as part of their job may be more nuanced. Hence, response options were anchored from 1 (strongly disagree) to 5 (strongly agree), such that higher scores indicated greater expectations that work-related activities on social media were an expected part of one’s job. Work-related social media use. This measure evaluated the frequency of employees’social media use for work-related communication and was derived from Van Zoonen et al. (2016). Respondents were prompted to respond to six items, such as “I publish work-related content on social media”by asking how often they engaged in these behaviors, from never (1) to multiple times a day (7). Controls. Additionally, gender, age, employment type, occupational role, and social media training were used as control variables. Gender was coded (1) female or (2) male, and employment type referred to whether employees had permanent (1) or fixed-term (2) contracts. Occupational roles represented in our sample were categorized as managerial roles (2) and other (1). Finally, social media training was measured for those who received training (2) and those who did not (1). Table A1 presents the description of scale items used in the analysis. Analysis Structural equation modeling was used to test the hypothesized models, and several fit indices were used to gauge model fit. Specifically, two incremental fit indices were used—namely, the Tucker–Lewis index (TLI) and the comparative fit index (CFI)—and two absolute fit indices—namely, a standardized version of the root mean square residual (SRMR) and the root mean square error of approximation (RMSEA). Additionally, the χ 2 statistic was reported. In order to estimate model parameters and corresponding confidence intervals, 5000 bootstrap samples were extracted from the data. RESULTS Measurement model The measurement model demonstrated good model fit: χ 2 (312) =1566.49; CFI =0.950; TLI =0.944; SRMR =0.04 and RMSEA =0.058 (CI: 0.056, 0.061). The results of the confirmatory factor analysis indicated no validity concerns. Specifically, we examined reliability through composite reliability scores and by calculating Hancock’s coefficient Hfor maximum reliability (Hancock & Mueller, 2001). The coefficient His a reliability measure where the squared correlation between a latent construct and the optimum linear composite is formed by its indicators. Distinct from other reliability measures (e.g., Cronbach’s alpha), His never less than the best indicator’s reliability, suggesting that a factor inferred from multiple indicators should never be less reliable than the best single indicator alone (see Hancock & Mueller, 2001 for a methodological explanation). The findings demonstrated that the composite reliabilities ranged between .83 and .95, whereas the maximum reliability (H) ranged between .87 and .95, which is well above the recommended threshold of .70. Subsequently, we examined convergent and discriminant validity. The findings showed that the average variance extracted (AVE) for all concepts in our model was well above the recommended .50, ranging between .55 and .74. All factor loadings were significant and sizeable on the intended latent construct, ranging between .63 and .96. Discriminant validity was also demonstrated, as the square root of the AVE was greater than the inter-construct correlations. Further, the maximum shared variance (MSV) ranged between .15 and .34. Discriminant validity was established as the MSV did not exceed the AVE for all the constructs in the model (Hair et al., 2010). Overall, these findings indicated adequate measurement of the model and justified further inspection of the structural relationship between these constructs. Table 1shows the validity and descriptive statistics. 6PEKKALA and VAN ZOONEN 17404762, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/emre.12597 by University Of Jyväskylä Library, Wiley Online Library on [18/08/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License Structural model Before testing the hypothesized model, we examined the variation in communicative role perceptions among employees. The mean (2.62) and median (2.75) were slightly below the mid-point of the 5-point scale. The variance was 0.93, and the data were normally distributed. The results further showed that 9.1% of the respondents agreed that these communicative behaviors were an expected part of their jobs, whereas 28.9% disagreed that the activities were expected job responsibilities. Overall, employees conceptualized communicative behaviors differently, varying in the extent to which they perceived work-related communication as an expected part of their job, with most employees centering around the scale’s mid-point (62% scored between 2.01 and 3.99). In the analyses, we controlled for gender, age, employment type, occupational role, whether someone had participated in social media training provided by the company, and for the company itself. The hypothesized model with control variables demonstrated good model fit: χ 2 (422) =1988.49; CFI =0.940; TLI =0.929; SRMR =0.04 and RMSEA =0.058 (CI: 0.056, 0.061). We found that older employees were less likely to engage in workrelated social media use (B=.118, BC95% [.169; .064] p=.001). Gender or employment type did not affect role perceptions or work-related social media use. Occupational role demonstrated strong relationships with both role perception (B=.346, BC95% [.225; .480] p=.001) and work-related social media use (B=.589, BC95% [.344; .854] p=.001). These findings indicated that those in managerial roles have broader role definitions and report higher levels of work-related social media use. Moreover, social media training increased role perceptions (B=.061, BC95% [.023; .099] p=.004) and work-related social media use (B=.256, BC95% [.193; .329] p=.001). These findings suggest that role perceptions and social media use may vary depending on the occupational role of employees and on whether they have had social media training. Finally, the company was found to significantly affect role perceptions and social media use (see Table 2). Notably, a comparison of means across the companies using a one-way ANOVA with Bonferroni correction demonstrated that role perceptions were lowest in company C (M=2.29, SD =0.92) compared with company A (M=2.64, SD =0.99) and company B ([M=2.80, SD =0.86]; Eta =.022 CI95% .008, .040). Social media use was highest in company B (M =3.07, SD =1.17) compared with company A (M=2.82, SD =1.39) and company C(M=2.98, SD =1.26). However, it should be noted that the results for the hypotheses did not change based on the inclusion or exclusion of these control variables (see Table 2). Hence, as we lack clear theoretical TABLE 1 Validity and descriptive statistics. Variable M(SD) CR AVE MSV MaxR(H) 1 2 3 4 5 6 7 8 9 10 1. Leadership Support 2.75 (0.92) .93 .74 .32 .95 .86 2. Impact 3.77 (0.85) .83 .63 .16 .87 .27 .79 3. Identification 3.95 (0.74) .90 .66 .15 .96 .29 .39 .81 4. Communicative role perceptions 2.62 (0.96) .95 .69 .34 .95 .56 .32 .19 .83 5. Work-related social media use 2.89 (1.34) .88 .55 .34 .89 .35 .40 .27 .58 .74 6. Gender a 1.40 (0.51) - - - - .04 .06 .04 .00 .00 - 7. Age n/a - - - - .10 .12 .10 .09 .01 .05 - 8. Employment type 1.09 (0.29) - - - - .01 .03 .05 .02 .06 .07 .23 - 9. Occupational role 1.12 (0.33) - - - - .13 .17 .19 .21 .26 .13 .20 .08 - 10. Social media training n/a - - - - .19 .29 .13 .27 .38 .01 .02 .04 .13 - Note: Square root of the AVE is reported on the diagonal. Correlations ±.06 are significant at p< .05. Abbreviations: CR, composite reliability; AVE, average variance extracted; MSV, maximum shared variance; ASV, average shared variance. a Coded 1 female 2 male. Employment type was coded 1 permanent contract 2 fixed term. Occupational role was coded 1 no managerial role, 2 managerial role. EMPLOYEES’COMMUNICATIVE ROLE PERCEPTION 7 17404762, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/emre.12597 by University Of Jyväskylä Library, Wiley Online Library on [18/08/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License TABLE 2 Regression results of structural models. Relationships Retained model Model with controls B SE Beta BC CI95% P B SE Beta BC CI95% P Direct effects (X–Y) Leadership support àsocial media use .022 .033 .036 .144; .077 .508 .021 .032 .020 .140; .078 .715 Impact àsocial media use .205 .029 .463 .328; .609 .001 .181 .028 .450 .277; .542 .001 Identification àsocial media use .090 .030 .206 .065; .338 .002 .064 .027 .168 .009; .266 .010 Direct effects (X–M) Leadership support àrole perceptions .519 .027 .534 .475; .593 .001 .486 .028 .524 .440; .560 .001 Impact àrole perceptions .199 .032 .277 .185; .362 .001 .167 .031 .258 .143; .317 .001 Identification àrole perceptions .037 .031 .053 .139; .032 .220 .056 .031 .073 .164; .008 .096 Direct effect (M–Y) Role perceptions àsocial media use .511 .034 .825 .700; .947 .001 .433 .035 .783 .581; .819 .001 Indirect effects (X–M–Y) Leadership àrole perceptions àsocial media use .428 .039 .441 .363; .520 .001 .341 .035 .351 .282; .422 .001 Identification àrole perceptions àsocial media use .031 .026 .043 .116; .026 .213 .039 .022 .056 .120; .004 .071 Impact àrole perceptions àsocial media use .164 .030 .229 .155; .316 .001 .117 .025 .164 .097; .235 .001 Control variables Gender àrole perceptions .031 .026 .078 .144; .024 .179 Age àrole perceptions .024 .024 .018 .021; .055 .351 Employment type àrole perceptions .004 .022 .013 .143; .123 .862 Occupational role àrole perceptions .115 .023 .346 .225; .480 .001 Training àrole perceptions .151 .026 .061 .023; .099 .004 Company A a àrole perceptions .053 .035 .102 .031; .233 .133 Company B b àrole perceptions .180 .035 .415 .263; .572 .001 Gender àsocial media use .020 .025 .056 .021; .055 .408 Age àsocial media use .074 .030 .118 .169; .064 .001 Employment type àsocial media use .073 .024 .360 .600; .129 .006 Occupational role àsocial media use .133 .028 .589 .344; .854 .001 Training àsocial media use .236 .027 .256 .193; .329 .001 Company A a àsocial media use .122 .036 378 .588;. -152 .003 Company B b àsocial media use .099 .037 .368 .648; .105 .006 Note: All results in the table were obtained by extracting 5000 bootstrap resamples from the data using a maximum likelihood estimator. B represents standardized coefficients, and SE represents the standard error associated with the standardized coefficient. BC CI95% represents the 95% bias-corrected confidence interval associated with the unstandardized coefficient (Beta). The p-values are associated with the unstandardized coefficients. a Dummy variable represents comparison of company A (1) versus company B and company C (0). b Dummy variable represents comparison of company B (1) versus company A and company C (0). 8PEKKALA and VAN ZOONEN 17404762, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/emre.12597 by University Of Jyväskylä Library, Wiley Online Library on [18/08/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License APPENDIX TABLE A1 Description of scale items used in analysis. Measurement Items Scales Perceived leadership support (adapted from Lewis et al., 2003) 1. My manager is committed to a vision of using social media at work. 2. My manager is committed to supporting my efforts in workrelated social media communication. 3. My manager encourages the use of social media at work. 4. My manager recognizes my efforts in using social media for work. 5. Employees’use of social media for work-related communication is important to my manager. Answers range from 1 “strongly disagree”to 5 “strongly agree” Perceived impact of communication (adapted from Spreitzer, 1995) 1. My communicative actions have an impact on my company’s business performance. 2. The way I communicate affects how our customers perceive us as a company. 3. My communicative actions have a significant influence on how my organization is perceived as an employer. Answers range from 1 “strongly disagree”to 5 “strongly agree” Organizational identification (adapted from Smidts et al., 2001, used, e.g., by Fieseler et al., 2015) 1. I feel strong ties with my organization. 2. I experience a strong sense of belonging to my organization. 3. I feel proud to work with my organization. 4. I am sufficiently acknowledged in my organization. 5. I am proud to work for my organization. Answers range from 1 “strongly disagree”to 5 “strongly agree” Communicative role perception (based on Morrison, 1994 and Van Zoonen et al., 2016) 1. I am expected to use social media in my work to network with new people relevant for my work. 2. I am expected to use social media in my work to tell others about my work. 3. I am expected to use social media in my work to maintain professional relationships. 4. I am expected in my work to create and publish professional content on social media. 5. I am expected to follow social media in my work to stay up to date with stakeholders’activities and industry news and events. 6. I am expected in my work to share my organization’s news and accomplishments on personal social media. 7. I am expected in my work to participate in work-related discussions on social media. 8. I am expected in my work to mention my employer’s name in my social media. Answers range from 1 “strongly disagree”to 5 “strongly agree” Work-related social media communication (derived from Van Zoonen et al., 2016) 1. I publish work-related content on social media. 2. I share work-related content created by others on social media. 3. I participate in discussions related to my work on social media. 4. I read other people’s posts on social media to obtain workrelated information and knowledge. 5. I invite people to join my social media network because I think they might be valuable professional contacts. 6. I maintain and strengthen relationships with colleagues, customers, and business partners by sending messages to people in my network. Answers range from 1 “never”to 7 “multiple times a day” EMPLOYEES’COMMUNICATIVE ROLE PERCEPTION 15 17404762, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/emre.12597 by University Of Jyväskylä Library, Wiley Online Library on [18/08/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License