Information and communication technologies (ICTs) are becoming an integral part of individual’s working life (Vranjes et al., 2018). The use of ICTs has been accelerated by the COVID-19 pandemic and is expected to stay pervasive beyond (Czakert et al., 2021). ICTs have played a great role in increasing the efficiency, effectiveness, and productivity of employees; however, scholars and practitioners alike have recently become cognizant of their downside as well (Anwar et al., 2020; Vranjes et al., 2021). It has become evident in the last few years that ICTs often serve as a platform for interpersonal mistreatment, such as cyberstalking, cyberharassment, and cyberbullying (Lazuras et al., 2017). Despite being a relatively new phenomenon, cyberbullying is now well recognized as a serious public health hazard affecting children, adolescents, and adults alike (Aboujaoude et al., 2015). Research on the topic of cyberbullying has typically been conducted in school settings since the early 2000s (Farley et al., 2016; Vranjes et al., 2021). However, recent research highlights that adult employees in the workplace are not insusceptible to cyberbullying (Loh & Snyman, 2020). Workplace cyberbullying (WCB) is increasingly been recognized as a significant problem facing modern organizations (Coyne et al., 2019). Regarding the prevalence of WCB, Forssell (2016) found that 9.7 % of the respondents could be classified as WCB victims, whereas Coyne et al. (2017) found that 20.8 % of the respondents could be regarded as WCB victims using Leymann’s cut-off criterion, that is, at least weekly exposure to negative acts during the past 6 months. While researchers know an increasing amount about workplace face-to-face (FTF) bullying, the extant literature on WCB is scarce (Farley et al., 2016). Although WCB occurs less frequently than workplace FTF bullying (Barlett et al., 2021); however, some researchers argue that exposure to WCB may have more negative repercussions for victims compared to workplace FTF bullying (Coyne et al., 2017) because certain features of WCB (e.g., anonymity, 24/7 accessibility, a potentially large audience, and online permanence) increase fear and uncertainty (Ford, 2013, D’Souza et al., 2018). Indeed, research has demonstrated that WCB victimization leads to a variety of negative personal and organizational outcomes, such as unfavorable job attitudes, withdrawal behaviors, and physical and mental health issues (Coyne et al., 2017). How organizations prevent and mitigate WCB is thus an important business ethics issue (Coyne et al., 2019), particularly in the COVID-19 era that has witnessed higher prevalence rates of WCB due to the increased use of the Internet among employees (Barlett et al., 2021).

Scant empirical research on WCB has predominately focused on the prevalence rates and its negative consequences for victims and organizations, including depression, anxiety, stress, negative job attitudes, interpersonal deviance, decreased task performance, and turnover intentions (e.g., Cassidy et al., 2014; Kowalski et al., 2018; Loh & Snyman, 2020; Park & Choi, 2019; Zhang et al., 2021). Also, some researchers have explored the personal and organizational determinants of WCB victimization, such as demographics, technology usage, work stressors, organizational climate, and organizational culture (e.g., Forssell, 2016, 2020; Iftikhar et al., 2021; Kim & Choi, 2021; Vranjes et al., 2018). However, only little is known about situational factors that may encourage individuals to become perpetrators of WCB (Vranjes et al., 2017). Recently, Vranjes et al. (2021) through a diary study demonstrated that men’s daily work stressors spilled over to their private life in the form of anger after work and antisocial online behavior throughout the evening. In another study, Zhang and Leidner (2018) found that perpetrators use three denial neutralization techniques (i.e., denial of responsibility, denial of injury, and denial of victim) to justify their WCB behavior.

Usually, bullying involves distinct bully and victim roles, but research has demonstrated that a ‘bully/victim’ role may also emerge. Bully/victims are usually individuals who engage in both bullying perpetration and victimization roles at different times (Vranjes et al., 2021). Bully/victims represent a small but distinct group in the bullying literature and tend to be more aggressive than pure bullies (Salmivalli & Nieminen, 2002), and have more negative affect than pure victims (Powell & Ladd, 2010). Research has revealed that bully/victims are unlikely to engage in aggressive and anxious behavior patterns if they are not first bullied themselves, thus underscoring the role of environment in the development of bully/victims. Role overlap is the tendency of victims to become perpetrators, and perpetrators to become victims (Lazuras et al., 2017; Smith, 2019). The role overlap hypothesis has already received support in the school context wherein researchers have found that victims of FTF bullying more frequently engage in both FTF bullying and cyberbullying than non-victims (e.g., Lazuras et al., 2017). However, the emerging WCB literature is yet to empirically test the role overlap hypothesis. Besides, Vranjes et al. (2017) noted that similar situational predictors may be of importance for the two roles, as both bullies and victims are subjected to the same organizational context. Thus, based on affective events theory (AET; Weiss & Cropanzano, 1996) and the Emotion Reaction model (ERM; Vranjes et al., 2017) of WCB, we propose a research model that posits that perceived organizational politics (POP) as a situational factor can trigger both forms of bullying (i.e., offline and online) in the workplace, and discrete negative emotions—anger and fear—play a key role in this process (see Fig. 1). Further, the model assumes that the two roles (i.e., bully and victim) may be swapped among victims and perpetrators. More specifically, we expect that in a politically charged work environment victims of workplace FTF bullying may become perpetrators of WCB subsequently. AET (Weiss & Cropanzano, 1996) proposes that features of the work environment influence the occurrence of affective events that engender affective reactions in employees, which in turn shape employees’ attitudes and behaviors. The ERM (Vranjes et al., 2017) that aligns with the main proposition of AET proposes that discrete negative emotions (e.g., anger, fear, and sadness) elicited by work stressors play a central role in the cyberbullying process.

Fig. 1
figure 1

Research model and hypotheses

Consistent with the main proposition of the ERM (Vranjes et al., 2017), we argue that depending on the emotions experienced an employee will be either on the delivering or receiving end of bullying behavior. Specifically, employees who will experience anger as a consequence of POP may engage in WCB perpetration as these employees are more likely to express their intense negative emotions more overtly online as opposed to face-to-face (Derks et al., 2008). Conversely, employees who will experience fear as a consequence of POP may become easy targets for workplace FTF bullying at the hands of those employees who hold power (formal and/or informal) in a politically charged work environment. In turn, employees who are victims of workplace FTF bullying may engage in WCB perpetration as the online environment may provide these otherwise powerless employees the opportunity to retaliate or get even against their more powerful face-to-face aggressors while possibly remaining anonymous (Vranjes et al., 2020). Finally, the research model shown in Fig. 1 also assumes that a favorable attitude toward WCB is the proximal predictor of WCB perpetration (Barlett et al., 2016).

We tested our hypotheses with data from a sample of 279 faculty members of public sector universities in Islamabad, Pakistan. Our contribution to the WCB literature is fourth-fold. First, the WCB literature that is still in its infancy has focused more on personal and organizational outcomes related to WCB victimization, yet relatively little is known about the situational antecedents of WCB perpetration. In the current study, we propose that POP as a situational factor may encourage cyberbullying in the workplace. We expect that discrete negative emotions of anger and fear will play a central role in linking POP to WCB perpetration. Second, it has been observed that in most situations cyberbullying happens alongside FTF bullying (Kim & Choi, 2021). Researchers have emphasized the importance of including both forms of bullying when investigating the bullying construct in the work context, as omitting either FTF bullying or cyberbullying might provide an incomplete picture of the situation (e.g., Vranjes et al., 2020). Research has also demonstrated that FTF bullying and cyberbullying are highly correlated (e.g., Kowalski et al., 2018). We thus respond to this call and integrate both forms of bullying in our conceptual model and expect that in a politically charged work environment FTF bullying victimization is likely to predict WCB perpetration. Third, to the best of our knowledge, this is the first study to empirically test the role overlap hypothesis in the work context. To this end, we propose that FTF bullying victims may change role and become bullies through cyberbullying perpetration to retaliate against their higher status face-to-face aggressors or to protect themselves from future victimization. Lastly, our study provides important insights into the process through which a favorable attitude toward cyberbullying is formed. We propose that positive attitude toward cyberbullying is the proximal predictor of WCB perpetration. We outline implications for managers and organizations based on our findings to make a case as to how organizations can prevent and mitigate WCB.

Theoretical Background and Hypotheses

AET focuses on the structure, causes and consequences of affective experiences in the workplace (Weiss & Cropanzano, 1996). This is a useful framework for understanding how POP predicts negative emotions, such as anger and fear, which in turn predict attitudinal and behavioral outcomes. The crux of the theory is that factors in the work environment that impede an employee’s progress towards workplace goals lead to affective responses (Ashton-James & Ashkanasy, 2008; Weiss & Cropanzano, 1996). A core proposition of AET is that characteristics of the work environment generate affective events that cause affective reactions in employees, which in turn, determine employees’ attitudes and behaviors (Weiss & Cropanzano, 1996). Much of the literature on affective events focuses on negative work conditions that elicit a stress response and a perceived threat to one’s work-related goals (Ashton-James & Ashkanasy, 2008; Lazarus, 1991; Vranjes et al., 2017), thereby leading to affective, attitudinal, and behavioral responses.

POP involves an individual’s attribution to behaviors of self-serving intent, and is defined as “an individual’s subjective evaluation about the extent to which the work environment is characterized by co-workers or supervisors who demonstrate such self-serving behavior” (Ferris et al., 2000, p. 90). An important aspect of POP is that it involves perceptions of illegitimate, self-serving behaviors and is thus viewed as a dysfunctional and divisive work condition (Chang et al., 2009; Hochwarter et al., 2020). POP is conceptualized to elicit a stress response (Cropanzano & Li, 2006; Ferris et al., 1996; Liu et al., 2006), is correlated with stress (Ferris et al., 1996; Miller et al., 2008) and strain (Chang et al., 2009; Hochwarter et al., 2020), and represents a perceived workplace threat (Ferris et al., 2002) that can mitigate individual goal attainment. POP and related constructs, such as political tactics and intimidation, have therefore been studied using AET (Rosen et al., 2009; Windsor, 2016). Previous research has leveraged AET specifically to explain the role of negative emotions as mediators of relationships between POP and negative attitudinal and behavioral outcomes (e.g., Liu et al., 2006; Rosen et al., 2009), and has conceptualized POP as a negative work condition that relates to an affective event in the form of negative emotions and, hence, to attitudinal and behavioral outcomes.

Consistent with AET and this stream of research, we propose that POP should elicit both anger and fear (Meisler, 2020b; Rosen et al., 2009; Thiel et al., 2014). Anger and fear should, in turn, relate to WCB perpetration and FTF bullying victimization, respectively. AET and research on this theory suggest that higher levels of POP should be related to negative emotions, such as anger and fear (Ashton-James & Ashkanasy, 2008; Liu et al., 2006; Rosen et al., 2009), which should in turn be related to negative behavioral outcomes, such as bullying behavior (Naseer et al., 2016). Integrating this perspective with the ERM (Vranjes et al., 2017) provides the theoretical rationale as to why anger and fear should be differentially related to WCB perpetration and FTF victimization. The ERM proposes that anger and fear mediate relationships between work-related stressors (e.g., POP) and cyberbullying behavior (Vranjes et al., 2017). Furthermore, the model proposes that anger links work stressors to bullying perpetration, whereas fear links work stressors to bullying victimization (Vranjes et al., 2017).

In the sections below, we build on propositions from AET and the ERM as well as research on these theories related to POP, in the development of our conceptual model and hypotheses.

Perceived Organizational Politics and Negative Emotions

Recently, organizational politics researchers have acknowledged the potential contribution of emotions to explaining the perceptions and implications of organizational politics (Meisler, 2020b). Research suggests that POP can provoke a broad range of positive and negative emotions in employees because different types of employees are likely to appraise, evaluate, and interpret POP differently (e.g., Liu et al., 2006; Basch & Fisher, 2000). The discrete emotion theory suggests that affective experiences can be reduced to a limited set of universal basic emotions, most commonly identified as happiness, sadness, anger, fear, and disgust (Stadthagen-González et al., 2018). In the present study, we view POP as a dysfunctional aspect of the work environment and assume that POP will provoke discrete negative emotions of anger and fear in employees. Anger is defined as “a response to threat that harms one’s goal,” whereas fear as “a response to threat, uncertainty, and danger to the self” (Xu et al., 2020, p. 4).

The behavior that comprises POP is behavior that is deemed to be self-serving and that occurs at the expense of others in the organization (Bedi & Schat, 2013). Examples of organizational politics include backstabbing, influence tactics, favoritism-based employment decisions, and self-promotion. Considering these examples, employees’ POP is generally seen as dysfunctional (Poon, 2003). POP represents a divisive work condition that threaten one’s goal attainment (Chang et al., 2009; Hochwarter et al., 2020). POP, therefore, is likely to be appraised as a dysfunctional and threatening aspect of the work environment because it not only thwarts employees’ personal and professional goal attainment, but also creates considerable ambiguity and unpredictability among employees by putting them at risk for losing the things that they have already obtained, such as their reputation, their power, and their status in the organization (Cropanzano et al., 1997). Thus, POP represents threats to one’s goal attainment as well as to the self; POP should therefore predict anger and fear, respectively. Having said that, employees may be more likely to experience anger or fear based on their personality (Flaa et al., 2007) and, in conjunction, the way in which they react to stressful work conditions (Carver & Connor-Smith, 2010). For instance, individuals who are more approach-oriented may be more likely to experience anger, whereas avoidance-oriented individuals may be more likely to experience fear in relation to stressful conditions, such as POP (e.g., Carver & Connor-Smith, 2010; Carver & Harmon-Jones, 2009).

AET (Weiss & Cropanzano, 1996) provides theoretical logic for the relationships we propose between POP, anger, and fear. POP represents a divisive and dysfunctional work condition (Chang et al., 2009; Hochwarter et al., 2020; Poon, 2003) that should stimulate a stress response (Cropanzano & Li, 2006; Ferris et al., 1996; Liu et al., 2006). AET proposes that work conditions that threaten one’s work-related goal attainment and elicit a stress response should relate to negative emotions, such as anger and fear (Cropanzano & Li, 2006; Lazarus, 1994; Liu et al., 2006). Research on AET suggests that work conditions perceived as threatening, such as POP, should be related to anger and fear (Ferris et al., 2002; Hochwarter et al., 2020). Research supports these arguments that POP and related constructs, such as political tactics and intimidation, elicit negative affective responses (Meisler et al., 2017; Meisler, 2020a; Thiel et al., 2014) including anger and fear specifically (Drory & Meisler, 2016; Meisler, 2020b; Rosen et al., 2009). Therefore, we hypothesize:


POP is positively associated with anger.


POP is positively associated with fear.

Perceived Organizational Politics and Workplace Face-to-Face Bullying Victimization

FTF bullying is a form of interpersonal mistreatment characterized by three elements: persistency, power imbalance, and intent to harm (Einarsen et al., 2020). First, persistency of the inappropriate behaviors of a primarily psychological nature in terms of repetition (at least once a week), duration (within a 6-month period), and patterning (of a variety of negative and aggressive behaviors involved) is one of the most salient elements of FTF bullying which differentiates it from single and isolated aggressive events (Einarsen & Skogstad, 1996; Leymann, 1996; Olweus, 1991). Thus, FTF bullying is often subject to escalation over time (Rai & Agarwal, 2018). Second, a power imbalance must exist between the perpetrator and the target, whereby the target finds it increasingly difficult to defend himself or herself (Einarsen et al., 1994). The power imbalance has been conceptualized as being derived from the perpetrator’s hierarchical position in the organization (e.g., supervisor), informal power based on knowledge and experience as well as access to support from influential persons, and the target’s social, physical, financial or even psychological dependency on the perpetrator (Einarsen et al., 2020). Third, negative acts must be systematic and premeditated, and the presence of negative intent on the part of the perpetrator is a key element of FTF bullying. Nonetheless, there is no general consensus in the bullying literature whether intent to harm should be a defining element of FTF bullying (Einarsen et al., 2020).

The notion that FTF bullying is largely attributable to deficiencies in the work environment is usually referred to as the ‘work environment hypothesis’ (Einarsen, 2000). Supporting this perspective, researchers have identified several organizational factors associated with FTF bullying, such as role stressors (Reknes et al., 2014), organizational culture (Pilch & Turska, 2015), organizational climate (Bond et al., 2010), leadership styles (Hoel et al., 2010), power and control (Ferris et al., 2007), and POP (Amponsah-Tawiah & Annor, 2017; Salin, 2003). Ferris et al. (2007) noted that POP creates an environment of uncertainty where employees feel that they do not have control. In other words, bullying is a manifestation of politically charged work environments (Ferris et al., 2007; Neuman & Baron, 2003).

Employees may choose to respond to POP through bullying to regain some control over their work environment. Further, employees in a politically charged work environment may feel that their organization permits and supports employees who are part of the strongest cabal thus facilitating higher incidents of workplace FTF bullying directed toward employees who are less powerful (Naseer et al., 2016). That being the case, perceptions of bullying victimization, in general, should be higher in work environments characterized by organizational politics. Thus, an individual’s POP should be positively related to bullying victimization. Supporting these arguments, Salin (2003) and Amponsah-Tawiah and Annor (2017) in their studies found a positive association between POP and FTF bullying victimization. Similarly, Naseer et al. (2016) demonstrated that POP positively predicts FTF bullying victimization, suggesting that organizational politics nourishes a stressful work environment in which less powerful employees are more likely to experience FTF bullying. Thus, we derive:


POP is positively associated with workplace FTF bullying victimization.

Anger and Workplace Cyberbullying Perpetration

WCB can be distinguished from workplace FTF bullying due to its unique features that are enabled by technology usage (Scisco, 2019). Some criteria regarded as the mainstay of FTF bullying (i.e., repetition and power imbalance) hold a different meaning in the virtual context (Vranjes et al., 2020). For instance, for FTF bullying to be repetitive, the perpetrator would engage in repetitive negative behaviors. On the other hand, one act of cyberbullying can be experienced repeatedly by the target and observers such as posting a harmful video online (which can then be viewed repeatedly) or re-reading an email (Dooley et al., 2009). Moreover, FTF bullying can be categorized by a power imbalance, with the perpetrator typically having more physical, psychological or social power than the target (Scisco, 2019). In contrast, a victim can be vulnerable to cyberbullying because the perpetrator may gain power by having advanced technological skills or by becoming anonymous by creating fake e-mail accounts, online IDs or phone numbers (Dooley et al., 2009; Vandebosch & van Cleemput, 2008). Further, unlike FTF bullying, there is potentially no reprise from technology-based interactions as they can occur anytime and anywhere (Farley et al., 2021). In this sense, the inability to have any control over the acts of bullying may leave the victim helpless and powerless (Vranjes et al., 2020). Accordingly, WCB can be defined as “negative, technology-mediated behaviors that are repetitively experienced by a vulnerable target within the work context” (Scisco, 2019, p. 83). Examples of WCB include nasty text messages or e-mails, rumors sent by e-mail or posted on social networking sites as well as embarrassing pictures, videos, websites, or fake profiles.

The ERM (Vranjes et al., 2017) posits that anger experienced as a consequence of workplace stressors potentially prompts an employee to engage in WCB perpetration. This is consistent with the basic tenet of the frustration-aggression hypothesis which states that aggression is a certain outcome of any frustration (Dollard et al., 1939). Anger is an outward-focused emotion that motivates other-blame and retaliation (Van Doorn et al., 2014), and is generally designed to express strong dissatisfaction and displeasure (Deffenbacher, 2011). Derks et al. (2008) in their review of studies regarding emotions in computer-mediated communication (CMC) concluded that intense negative emotions are expressed more overtly in CMC as opposed in FTF communication, because it is likely to reduce negative social appraisal and the fear of isolation (e.g., Ho & McLeod, 2008; Manstead & Fischer, 2001). Supporting their conclusion, Vranjes et al. (2021) demonstrated that men who experienced work stressors during the day felt angry after work and engaged in cyberbullying perpetration throughout the evening to vent their work-related anger. Thus, engaging in cyberbullying might provide people a way to vent their negative feelings (Pabian & Vandebosch, 2014). AET (Weiss & Cropanzano, 1996) also emphasizes that negative emotions, such as anger, evoked by dysfunctional aspects of the work environment may lead to undesirable behaviors (e.g., WCB perpetration). Thus, we expect:


Anger is positively associated with WCB perpetration.

Fear and Workplace Face-to-Face Bullying Victimization

Einarsen (1999) has suggested that workplace bullying can be divided into two different kinds, namely dispute-related bullying and predatory bullying. Dispute-related bullying is triggered by work-related conflicts that escalate into a bullying situation. On the other hand, in predatory bullying, the victims may be bullied because certain personal characteristics and attributes (such as being afraid, scared, and nervous) make them an easier target for the bully (Glasø et al., 2007). According to Finkelhor and Asdigian (1996), these personal characteristics of individuals are likely to increase vulnerability to victimization because these characteristics have congruence with the needs, motives or reactivities of bullies. Employees who feel nervous, depressed, and insecure seem to signal to other employees that they will not retaliate or fight back if attacked (Powell & Ladd, 2010).

The ERM (Vranjes et al., 2017) explains why employees who experience negative emotions, especially fear, are more prone to become targets of bullying behavior specifically as a form of displaced aggression. Based on the work environment hypothesis, we have argued above that bullying is a manifestation of work environment deficiencies (Einarsen, 2000) including politically charged work environments (Ferris et al., 2007; Neuman & Baron, 2003). According to the ERM, individuals in such work environments will look to displace their aggression as a way to escape the deficiency in the work environment (Vranjes et al., 2017). Furthermore, this model proposes that fear makes individuals “easy targets” (p. 329) of displaced aggression (bullying) because it is perceived to be a sign of weakness among perpetrators, retribution is perceived to be less likely. In a similar vein, victim precipitation theory suggests that employees who display negative emotions, such as fear and anxiety, tend to be perceived by others as weak, vulnerable, and less able to defend themselves. The manifestation of anxiety or tension in their exchanges with others may provoke bullying behavior (Samnani & Singh, 2016). Supporting these arguments, Rodríguez-Muñoz et al. (2015) using a two-wave longitudinal design demonstrated that Time 1 anxiety was positively associated with Time 2 FTF bullying victimization. Likewise, Balducci et al. (2011) found that individuals who scored high on neuroticism were more likely to become victims of FTF bullying than those who scored low. Thus, we expect:


Fear is positively associated with workplace FTF bullying victimization.

Negative Emotions and Attitude Toward Workplace Cyberbullying

According to Eagly and Chaiken (1993), an “attitude is a psychological tendency that is expressed by evaluating a particular entity with some degree of favor or disfavor” (p. 1). This definition includes the three key features of attitudes, namely tendency, entity or attitude object, and evaluation. AET (Weiss & Cropanzano, 1996) posits that emotions that flow from affective work events ultimately shape employees’ attitudes and behaviors. According to Agnew (1992), anger increases the individual’s level of felt injury, creates a desire for revenge, energizes the individual for action, and lowers inhibitions. Anger researchers have observed that angry/frustrated individuals compared to their non-angry counterparts have a tendency to view the world as a dangerous or threatening place, where others cannot be trusted and are perceived as out to harm them (Deffenbacher, 2011). They tend to interpret the actions of others as motivated by hostile intent, as unjustified, and as blameworthy. While in an angry state, individuals tend to adopt attitudes and beliefs that are in line with their feelings (Brezina, 2010). Therefore, it is argued that angry individuals are more likely to view aggressive behavior as justifiable and develop a positive attitude toward aggression (Bernard, 1990). Supporting this idea, Brezina (2010) in their study found that anger was positively associated with attitudes favoring aggression, and part of the effect of anger on aggression was indirect, operating through attitudes favoring aggression. Researchers have long viewed anger as associated with fight responses and fear with flight responses (Cannon, 1932). However, some researchers view fear as a potential motivation for aggression (e.g., Blanchard & Blanchard, 1984; Simunovic et al., 2013). Specifically, fear-induced defensive aggression is likely to occur in a threatening situation where escape has been attempted but is not possible (Moyer, 1968). When individuals are faced with an uncontrollable or inescapable stressful situation, they are likely to adopt attitudes that may help them to avert the danger. It is worth noting that we did not find any empirical evidence indicating that fear may lead to increases in aggressive attitudes among working individuals. Taken together, in the current study, we assume that POP-induced negative emotions of anger and fear may lead employees to develop a favorable attitude toward WCB. Overall, we conclude:


Anger is positively associated with a favorable attitude toward WCB.


Fear is positively associated with a favorable attitude toward WCB.

Workplace Face-to-Face Bullying Victimization and Workplace Cyberbullying Perpetration

Research suggests that FTF bullying and cyberbullying can co-exist in the workplace (Vranjes et al., 2020). Workplace FTF bullying and WCB are highly correlated (Farley et al., 2016; Kowalski et al., 2018), highlighting the need to integrate both forms in theoretical models when investigating the bullying construct in the workplace (Vranjes et al., 2020). Further, research suggests that the two roles (i.e., bully and victim) may be swapped among victims and perpetrators (Smith, 2004, 2019), and it is highly likely that aggressive retaliation may occur in the context of FTF bullying, whereby victims become bullies themselves to fight back against their aggressors (Choi & Park, 2019). However, aggressive retaliation in the offline world requires certain capabilities and resources such as physical power, peer support, and self-efficacy and not all bullying victims have access to and can utilize these resources for retaliation purposes (Lazuras et al., 2017). Under such circumstances, victims of FTF bullying are likely to retaliate against their offline aggressors through cyberbullying perpetration (König et al., 2010). Indeed, the opportunity to remain anonymous, the viral reach, and the intrusive nature of the online environment gives power to less powerful individuals to retaliate against their more powerful offline aggressors (Vranjes et al., 2020). Supporting this argument, Forssell (2016) found that individuals in a supervisory role were more often exposed to WCB as opposed to individuals working in non-supervisory positions. In this study, we except that in a politically charged work environment, victims of FTF bullying are likely to fight back against their higher status offline aggressors through cyberbullying perpetration, possibly anonymously so that their retaliatory behaviors cannot be traced backed to them, thus reducing the likelihood of counter-retaliation. Hence, we conclude:


Workplace FTF bullying victimization is positively associated with WCB perpetration.

Workplace Face-to-Face Bullying Victimization and Attitude Toward Workplace Cyberbullying

The notion of a cycle of violence (Widom, 1989), wherein exposure to violence incites violence in the victim, somehow supports the overlap between victimization and perpetration in the bullying phenomenon. Averdijk et al. (2016) argue that victimization and the accompanying blow to one’s self-esteem, desensitization, and loss of status may trigger a positive attitude toward the use of violence. The authors further contend that victimization influences the appraisal and reward parameters (such as the cost–benefit analysis) of acting violently in subsequent situations in such a manner that victims become more attracted toward the benefits of violence perpetration than toward its costs. Moreover, the social learning theory (Bandura, 1986) may explain the extent to which participation in violent incidents, even as a bystander, pushes the victims to learn the power of bullying, and increases expectations of positive personal rewards if the victim could become a perpetrator (Falla et al., 2022). In this study, we posit that in a politically charged work environment, victims of face-to-face bullying may perceive bullying behavior as legitimate and justifiable and eventually become perpetrators to retaliate against their more powerful aggressors. However, victims of face-to-face bullying may not possess power in the real world to engage in offline bullying; consequently, they may develop a positive attitude toward cyberbullying because the anonymity inherent in the online environment makes it easier for these victims to take revenge against their offline bullies. Moreover, based on the social learning perspective, we posit that when the bullying behavior of powerful individuals is positively reinforced in a political work environment, victims are likely to learn the power of bullying and, consequently, they may develop a favorable attitude toward bullying, particularly cyberbullying perpetration due to the (perceived) anonymity offered by the Internet and mobile phones and the associated (lower) risk of being caught (Pabian & Vandebosch, 2014). From this standpoint, their tendency toward bullying behavior may be viewed as a rational choice to deliberately improve their own position by sabotaging or threatening their former (traditional) bullies to compete for access to scarce resources in an uncertain work environment. Thus, we derive:


Workplace FTF bullying victimization is positively associated with a favorable attitude toward WCB.

Attitude Toward Workplace Cyberbullying and Workplace Cyberbullying Perpetration

Individuals who view aggressive behavior as acceptable and justifiable are more likely to engage in aggression than individuals who view aggressive behavior as unacceptable and unjustifiable (Douglas & Martinko, 2001). Indeed, research has demonstrated that attitude toward aggression positively predicts aggressive behavior (e.g., Seddig & Davidov, 2018). To the best of our knowledge, no study has investigated whether attitude toward cyberbullying is a potential risk factor for cyberbullying perpetration in the work-related context. However, research conducted on schoolchildren and adolescents has demonstrated that attitude toward cyberbullying positively predicts cyberbullying behavior, suggesting that cyberbullies have more favorable attitudes toward cyberbullying than non-cyberbullies. For example, Barlett (2015) using a four-wave longitudinal design demonstrated that cyberbullying attitude was a very important risk factor for later cyberbullying behavior. In another study, Doane et al. (2014), found that attitude toward cyberbullying was the most robust predictor of cyberbullying intentions and cyberbullying perpetration. Drawing on AET (Weiss & Cropanzano, 1996), we expect that a positive attitude toward WCB is the proximal predictor of WCB perpetration, leading us to the following hypothesis:


Favorable attitude toward WCB is positively associated with WCB perpetration.


Participants and Procedure

Participants were full-time faculty members (including lecturers, assistant professors, and associate professors) working at different academic departments of three large public sector universities in Islamabad, Pakistan. Data were collected during the months of September through November in the Fall semester of 2021. For the first time since the outbreak of the COVID-19 pandemic in Pakistan in March 2020, all educational institutions, including universities, restarted academic activities using a hybrid mode of teaching in Fall 2021. At that time, the 4th wave of COVID-19 had subsided in Pakistan and the government decided to lift most of the COVID-related restrictions across the country. We collected time-lagged (i.e., two waves) and self-reported data by means of a self-administered questionnaire delivered personally during working hours in both waves of data collection. At time 1, 600 faculty members were approached and were asked about their demographics, POP, experience of negative emotions (i.e., anger and fear), and FTF bullying victimization. Participants also completed the Marlowe-Crowne Social Desirability Scale at time 1. We received back 367 completed questionnaires at the end of time 1. About 1 month later, at time 2, those 367 faculty members were approached again to provide responses regarding their attitude toward WCB and WCB enactment. At the end of time 2, we received back 279 completed questionnaires. We matched the responses of time 1 and time 2 with the key generated by each participant according to the guidelines provided in the questionnaires. That is, the participants were asked to provide their initials followed by the month of their birth. The statistical analyses are thus based on 279 completed set of responses, for a response rate of 46.5 %. In terms of gender, 57 % of the participants were male; the mean age of the sample was 38.23 years (SD = 2.45). On average, the participants were in their current job for 10.25 years; rank-wise about 46 % of the participants were lecturers, 32 % were assistant professors, and 22 % were associate professors. Since there is a limited number of professors and it is difficult to request their participation, thus no single professor participated in this study.


Perceived Organizational Politics

POP was measured using the 12-item Perceptions of Organizational Politics Scale (POPS) developed by Kacmar and Ferris (1991). A sample item is “Favoritism, rather than merit, determines who gets good raises and promotions around here.” Scale anchors ranged from 1 (strongly disagree) to 5 (strongly agree). Composite reliability (CR) was 0.956.

Negative Emotions

Following Vranjes et al. (2018), discrete negative emotions of anger and fear were measured with a selection of the Positive and Negative Affect Schedule—Expanded Form (PANAS-X) developed by Watson and Clark (1994). Anger and fear were measured with three items each. Sample items included “My job made me feel angry” and “My job made me feel frightened” for anger and fear, respectively. Participants were asked to indicate the extent to which they felt negative emotions at work during the last 30 days using a 5-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree). CR was 0.926 for anger and 0.869 for fear. The results of exploratory factor analysis indicated that two factors, rather than one factor, represented the six items best.

Workplace Face-to-Face Bullying Victimization

Workplace FTF bullying victimization was measured with the self-labeling approach using the following definition that was presented to the respondents (Einarsen & Skogstad, 1996, p. 191): “Bullying takes place when one or more persons systematically and over time feel that they have been subjected to negative treatment on the part of one or more persons, in a situation in which the person(s) exposed to the treatment have difficulty in defending themselves against them. It is not bullying when two equally strong opponents are in conflict with each other.” This definition was immediately followed by the instruction: “Using the above definition, please indicate whether or not you have been bullied at work over the last 6 months.” Scale anchors ranged from 1 (no) to 5 (yes, many times a week). This single question has been shown to be a valid measure of exposure to bullying at work (Nielsen et al., 2009).

Attitude Toward Workplace Cyberbullying

Attitude toward WCB was assessed with a 5-item scale developed by Barlett et al. (2016), adapted to the work-related context. A sample item is “It is alright to send harmful online messages/posts to someone at work.” Scale anchors ranged from 1 (strongly disagree) to 5 (strongly agree). CR was 0.888.

Workplace Cyberbullying Perpetration

WCB perpetration was measured with a 3-item scale developed by Ybarra et al. (2007), adapted to the work-related context. A sample item is “In the last year, how many times did you send rude or nasty comments to someone at work while online.” Scale anchors ranged from 1 (never) to 5 (everyday/almost everyday). CR was 0.840.

Control Variables

Self-reports of WCB perpetration may be sensitive to social desirability response bias, thus we included social desirability response tendency as a control variable. This construct was measured using a short 13-item version of the Marlowe-Crowne (Crowne & Marlowe, 1960) Social.

Desirability Scale. A sample item is “I am always courteous, even to people who are.

disagreeable.” The instrument was answered using a True/False format. CR was 0.70.

Moreover, we controlled for employees’ gender and time spent on digital media as they are related with WCB perpetration (Vranjes et al., 2021). Gender was dummy coded (0 = female, 1 = male), whereas time spent on digital media during a typical day was measured on a 5-point Likert scale with scores ranging from 1 (none) to 5 (6 h or more).

Analytical Strategy

Partial least squares structural equation modeling (PLS-SEM) was used to validate the research model. PLS-SEM has several strengths that made it appropriate for this study, including its soft distribution assumptions, its ability to handle complex research models that comprise many constructs and relationships, and its suitability for testing a theoretical framework from a prediction perspective (Benitez et al., 2020; Hair et al., 2019). To perform PLS-SEM, we used the SmartPLS v. 3.3.3 software (Ringle et al., 2015). As suggested by Hair et al. (2019), a two-step approach to data analysis was adopted. The measurement model was evaluated in the first step and the structural model was evaluated in the second step. The assessment of the measurement model involved examining the factor loadings, internal consistency reliability, convergent validity, and discriminant validity. The assessment of the structural model involved examining the statistical significance and relevance of the path coefficients, the coefficient of determination (R2), effect size (f2), and the predictive relevance (Q2) based on the blindfolding procedure. In addition, we assessed the model’s out-of-sample predictive power (Q2predict) by using the PLSpredict procedure (Shmueli et al., 2016). We used PLSpredict with 7 folds, with the aim of meeting the minimum sample size of N = 30 for the holdout sample (Hair et al., 2020), repeating this procedure 10 times. We compared the root mean squared error (RMSE) values of the PLS-SEM analysis with the linear regression model (LM). If the PLS-SEM analysis, compared to the naïve LM benchmark, yields lower prediction errors in terms of RMSE for all indicators of the key target construct, this indicates that the model has high out-of-sample predictive power (Hair et al., 2019). The bootstrap method (1,000 resamples) was used to generate standard errors, t-statistics, and confidence intervals to evaluate the statistical significance of the path coefficients (Hair et al., 2021).


Descriptive and Correlation Analysis

Table 1 shows the means, standard deviations, and intercorrelations among the study variables. The means of workplace FTF bullying victimization (self-labeled) and WCB perpetration were rather low (ranging from 2.197 to 2.380, respectively) pinpointing floor effect/variance restriction (Sischka et al., 2020). Using Leymann’s cut-off criterion of workplace bullying, that is, at least weekly exposure to negative acts during the last 6 months 9.3 % (n = 26) of the respondents in the current study could be regarded as victims of workplace FTF bullying. We performed a one-way ANOVA to determine if there were significant differences in the mean scores of WCB perpetration between the three academic ranks. A one-way ANOVA revealed that there was a statistically significant difference in the mean scores between at least two academic ranks [F (2, 276) = 24.298, p < 0.05]. The Bonferroni post-hoc test for multiple comparisons showed that the mean score for lecturers (Mean = 2.667, SD = 0.691) was statistically different from assistant professors (Mean = 2.189, SD = 0.671) as well as from associate professors (Mean = 2.062, SD = 0.568) at α = 0.05. However, no statistical difference in the mean scores was found between assistant professors and associate professors. The pattern of correlations between the study variables was in the expected direction. Table 1 also shows that the only control variable that was significantly related to the dependent variable (i.e., WCB perpetration) was time spent on digital media.

Table 1 Means, standard deviations, and correlations among constructs

Common Method Bias

We used two statistical techniques to rule out the possibility of common method bias. First, we conducted the Harman’s single-factor test (Podsakoff et al., 2003) and found that the indicators did not significantly load onto one single factor but rather five different factors. The first factor only explained 36.53 % of the variance in the indicators. Second, the full collinearity test suggested by Kock (2015) was used and the results indicated that the VIF values of all latent constructs were lower than 3.3 ranging from 1.091 to 2.288, suggesting that common method bias was not a significant problem with regard to our data.

Confirmatory Factor Analysis

The results of the confirmatory factor analysis are summarized in Table 2. Internal consistency reliability was demonstrated as the CR estimate for each scale was greater than the threshold value of 0.7 ranging from 0.840 to 0.956. To assess convergent validity, we assessed the factor loadings of scale items on their respective constructs ae well as the average variance extracted (AVE) value of each construct. Results indicated that most of the factor loadings were higher than the suggested threshold value of 0.7. One item of the POPS (i.e., POP6 “Don’t speak up for fear of retaliation”) that had a factor loading lower than 0.7 but higher than 0.6 was retained because it was important to the relevant factor, whereas two items (i.e., FEAR2 “My job made me feel afraid” and ATD5 “I have no reservations about using technology to hurt others when they deserve it”) that had factor loadings lower than 0.5 were omitted. As shown in Table 2, all the AVE values exceeded the cut-off value of 0.5 ranging from 0.637 to 0.807 (Hair et al., 2019; Legate et al., 2021). Discriminant validity of the constructs was demonstrated using two different techniques. First, as shown in Table 1, the square root of each construct’s AVE value was greater than its highest correlation with any other construct (Fornell & Larcker, 1981). Second, we used the heterotrait-monotrait (HTMT) ratio of correlations as proposed by (Henseler et al., 2015). As shown in Table 3, all the HTMT values were lower than the cut-off value of 0.85, lending support for discriminant validity.

Table 2 Confirmatory factor analysis results
Table 3 Discriminant validity (HTMT criterion)

Hypothesis Testing

As shown in Fig. 2, POP was significantly and positively associated with anger (β = 0.372; t = 9.706; p < 0.01), fear (β = 0.288; t = 5.575; p < 0.01), and workplace FTF bullying victimization (β = 0.281; t = 6.322; p < 0.01), lending support to H1a, H1b, and H2. Further, anger was significantly positively associated with WCB perpetration (β = 0.395; t = 12.654; p < 0.01), whereas fear was significantly positively associated with workplace FTF bullying victimization (β = 0.589; t = 16.111; p < 0.01), providing support to H3 and H4. Anger was significantly positively associated with attitude toward WCB (β = 0.252; t = 3.991; p < 0.01), lending support to H5a. However, the relationship between fear and attitude toward WCB was not significant (β = − 0.105; t = 1.400; p > 0.05), thus H5b was not supported. The results further indicated that workplace FTF bullying victimization was significantly and positively associated with WCB perpetration (β = 0.157; t = 2.841; p < 0.05) and attitude toward WCB (β = 0.413; t = 7.432; p < 0.01), lending support to H6 and H7. Finally, the results demonstrated that attitude toward WCB was significantly positively related to WCB perpetration (β = 0.434; t = 11.068; p < 0.01), thereby providing support to H8. The results of hypothesis testing are summarized in Table 4.

Fig. 2
figure 2

Results of hypothesis testing.**p < 0.01

Table 4 Hypothesis testing results

To assess the quality of the structural model, R2 values, f2 effect sizes, and the blindfolding-based Q2 values were computed for all the endogenous latent variables in the model. As shown in Fig. 2, 61.9 % of the variance in the criterion variable (i.e., WCB perpetration) was explained by anger, workplace FTF bullying victimization, and attitude toward WCB, indicating model’s considerable in-sample explanatory power. In terms of effect size, attitude toward WCB (f2 = 0.372), followed by anger (f2 = 0.308) were found to be the most significant predictors of WCB perpetration, depicting large and medium effect sizes, respectively. On the other hand, workplace FTF bullying victimization (f2 = 0.045) had a small effect size (Cohen, 1992). The Q2 values for all the endogenous latent variables obtained using the blindfolding procedure were larger than zero, demonstrating the predictive accuracy of the structural model. Lastly, the results of the PLSpredict procedure indicated that the Q2predict values of all indicators of the endogenous constructs were positive. Likewise, the Q2predict values of the endogenous constructs were also positive. Moreover, for the three indicators of the key target construct, that is, WCB perpetration, we found that the RMSE values for the PLS-SEM model were lower than that of the naïve LM benchmark. Thus, we concluded that the model has high out-of-sample predictive power (Hair et al., 2019).


Based on AET (Weiss & Cropanzano, 1996) and the ERM (Vranjes et al., 2017), we developed and tested a conceptual model that posited that POP (a divisive work condition) encourages cyberbullying behavior in the workplace, and discrete negative emotions (i.e., anger and fear) play an important role in this process. Our results demonstrated that POP can provoke discrete negative emotions of anger and fear among organizational members. Supporting the work environment hypothesis, results revealed that POP was positively associated with FTF bullying victimization, highlighting that a politically charged work environment creates conditions that encourage bullying behaviors (Naseer et al., 2016). Further, results showed that employees who experienced anger as a consequence of POP are more likely to engage in cyberbullying perpetration to vent their anger. Previous research suggests that individuals express their negative emotions more overtly online as opposed to offline (e.g., Derks et al., 2008). On the other hand, employees who experienced fear as a consequence of POP are more likely to become victims of FTF bullying at the hands of higher status individuals. This finding is in line with past research which suggests that employees who feel nervous, anxious, and depressed in the workplace seem to signal to other employees that they will not retaliate or fight back against their aggressors and, consequently, such employees become easy targets of FTF bullying perpetrated by those who hold more power gained from either formal or informal sources (e.g., Rodríguez-Muñoz et al., 2015). Moreover, our results indicated that employees who experienced anger are likely to develop a positive attitude toward WCB, suggesting that angry employees tend to interpret the actions of others as motivated by hostile intent, as unjustified, and as blameworthy and, consequently, they view aggressive behavior as acceptable and justifiable (Brezina, 2010). While individual differences may have resulted in some individuals being more likely to experience anger or fear in relation to POP (Carver & Connor-Smith, 2010; Carver & Harmon-Jones, 2009; Flaa et al., 2007), many respondents in our study might have experienced both anger and fear as we observed a significant positive correlation between anger and fear (Pearson’s r = 0.595). What is important to note here is that these affective reactions that occur in connection to POP are differentially predictive of bullying perpetration versus bullying victimization. Contrary to our expectations, however, fear was not significantly associated with attitude toward WCB. This finding may suggest that employees who display fear are more likely to engage in avoidance behavior and risk aversion, and such employees may only view aggressive behavior as acceptable when the threatening situation becomes inescapable. Another important finding of the study is that victims of FTF bullying may retaliate or fight back against their more powerful offline aggressors by engaging in cyberbullying perpetration. This finding suggests that the possibility to remain anonymous, the viral reach, and the intrusive nature of the online environment provides power to otherwise powerless individuals to retaliate against their aggressors who hold more power in the real world (e.g., Forssell, 2016; Vranjes et al., 2020). Results also revealed that when victims of FTF bullying perceive that engaging in bullying behavior may help in improving one’s own position in a politically charged work environment (Salin, 2003), they are likely to develop a favorable attitude toward bullying. Finally, our results showed that favorable attitude toward cyberbullying is positively associated with cyberbullying perpetration.

This study contributes to the extant literature in several ways. First, in the current study, we identified a situational factor (i.e., POP) that sets the stage for bullying behaviors (offline and online) in the workplace, and highlighted the central role of negative emotions in linking POP to its distal outcome (i.e., WCB perpetration). Specifically, our study emphasized that individuals who experience anger as a consequence of POP are likely to develop a positive attitude toward cyberbullying and vent their anger and frustration by engaging in cyberbullying perpetration. On the other hand, individuals who experience fear and anxiety as a consequence of POP are more likely to become victims of FTF bullying. These victims may then develop a favorable attitude toward cyberbullying and engage in cyberbullying behavior in an attempt to retaliate against their more powerful offline aggressors. Thus, discrete negative emotions (i.e., anger and fear) play a vital role in linking POP with WCB perpetration through two distinct mechanisms. Second, we responded to the call to integrate both forms of bullying—offline and online—in theoretical models when investigating the bullying construct in the workplace (e.g., Vranjes et al., 2020). In doing so, our results underscored that FTF bullying victimization is an important risk factor for the development of later cyberbullying behavior in the workplace. Third, in our conceptual model, we focused both on bullying victimization and enactment as bullies and victims work together as a system, and without examining the roles of each, it is difficult to understand the whole process of bullying in the workplace. Fourth, supporting the role overlap hypothesis, we demonstrated that in a politically charged work environment victims of FTF bullying may swap roles and become perpetrators of cyberbullying to retaliate against their more powerful offline aggressors, as the anonymity inherent in many forms of computer-mediated interactions may reduce the possibility of counter-retaliation. Lastly, there has been relatively little attempt in the WCB literature to unfold how the attitude toward cyberbullying is formed and explore its role in predicting cyberbullying behavior. Toward this end, we demonstrated that both experiencing anger in response to POP and being a victim of FTF bullying may play important roles in shaping individuals’ attitude toward WCB. Moreover, favorable attitude toward cyberbullying not only positively predicted cyberbullying perpetration but also emerged as its strongest predictor.

Given that cyberbullying results in a host of negative outcomes for employees and organizations (Coyne et al., 2017), it is important that organizations prevent and mitigate this topical workplace issue (Coyne et al., 2019). The results of this study offer several ways by which to do so. First, the key role that POP plays in provoking negative emotions in employees and promoting both offline and online bullying in the workplace warrants special attention. Knowing that POP stems from lack of transparency and fairness that undermines the effort-reward expectancy, strategies that organizations can employ to minimize POP include: (1) discouraging self-serving behaviors within their ranks, (2) ensuring that decision-making processes are not dominated by hidden agendas and long-term organizational well-being takes precedence over short-term personal gains, (3) establishing clear rules and guidelines to ensure fairness in the resource allocation process, (4) improving employees’ perceptions of fairness of performance appraisal processes, (5) ensuring that important organizational rewards (e.g., pay, promotion, and recognition) are strongly linked to performance, and (6) encouraging employees to speak out without any fear of personal negative consequences (e.g., De Clercq & Belausteguigoitia, 2017; Malik et al., 2019). Second, as FTF bullying victimization was found to be a risk factor for cyberbullying perpetration, thus interventions to prevent and reduce bullying in general might be promising in inhibiting or decreasing FTF bullying as well as cyberbullying enactment in the workplace. Toward this end, it is vital that leaders cultivate a workplace culture that is fair and equitable and in which every employee is treated with respect and dignity so that bullying behaviors do not become entrenched (Cleary et al., 2010). Further, an anti-bullying policy should include a clear definition of bullying and examples of undesirable behaviors as well as a clear statement that any kind of bullying is intolerable (Zapf & Vartia, 2020). Managers should also encourage the reporting of bullying incidents and ensure that bullying policies are adhered to fairly and in a timely and appropriate manner (Cleary et al., 2010). In connection, organizations should consider adopting reporting systems for cyberbullying, such as a helpline that can be used during both work and nonwork hours (Holmes, 2017). Third, as anger was found to be a predictor of cyberbullying perpetration, thus interventions aimed at anger reduction might be effective in decreasing cyberbullying incidents. Of the several treatment methods for anger reduction, variations of cognitive-behavioral therapy (CBT) are most common. These include inductive social skills training, skill assembly social skills training, and cognitive relaxation coping skills. Research has demonstrated that all three approaches reduce a variety of anger types including general anger and anger across a variety of situations (e.g., Deffenbacher et al., 1994). Employee assistance programs (EAPs) can help employees with negative emotions and mood disorders as well as with experiences of cyberbullying specifically (Chen et al., 2021). Fourth, our results demonstrated that individuals who tend to be depressed and anxious are more likely to become victims of FTF bullying and may subsequently seek cyber revenge against their aggressors. Interventions such as mindfulness-based trainings targeted at vulnerable individuals might be effective in reducing their negative affect and enhancing emotion regulation, thereby reducing their likelihood of becoming an easy target of bullying (Vranjes et al., 2018). Lastly, organizations should organize workshops to increase employees’ digital literacy and educate them about the “netiquette” (Brown, 2014, p. 285), including teaching employees how to avoid ambiguous communication and when to use which medium (e.g., email or telephone) for effective communication (Vranjes et al., 2021).

This study has some limitations that provide avenues for future research. First, our research model was tested with cross-sectional data. Thus, causation is inferred primarily from the theoretical lens of AET and the ERM. Future research might use longitudinal research designs (e.g., the cross-lagged panel model) to test causal processes or the possibility of reversed and/or reciprocal causation. However, given the lack of research into WCB, particularly the role of situational factors in predicting WCB perpetration, this study still offers valuable information into the phenomenon. Second, in the current study, we exclusively focused on the role of single situational factor (i.e., POP) in triggering WCB. Future studies may explore the role of other organizational factors (e.g., dysfunctional leadership, perceived injustice, and conflicts and communication problems) in predicting WCB perpetration. Moreover, future research may investigate the moderating roles of empathy and self-esteem as well as personality traits (e.g., the Light Triad traits) in attenuating the effects of anger and attitude toward cyberbullying on cyberbullying enactment. Third, we slightly adapted the existing cyberbullying attitude and cyberbullying perpetration scales to make them more compatible and suitable for the work-related context. It is worth noting that researchers have developed and tested scales to measure WCB victimization, however, to our best knowledge, no scale has yet been developed to gauge WCB perpetration (e.g., Escartín et al., 2021). Thus, future research may develop and test WCB scales that focus exclusively on perpetrators as well as bystanders as bystanders can play a vital role in addressing and managing cyberbullying in the workplace. Fourth, data in this study were gathered from faculty members working in public sector universities of Pakistan, which might have implications for the generalizability of our findings in the sense that there is not so much cross-cultural variability in POP (Vigoda, 2001), but in relations between POP with its correlates (Chang et al., 2009). Since expression of negative emotions, such as anger, is higher in more individualistic cultures (Matsumoto et al., 2010) and since Pakistan is more collectivistic than most Western cultures (Nadeem & de Luke, 2020), the pattern of results demonstrated in our study is even more likely in more individualistic, Western cultures. Future studies should thus replicate our findings in more diverse samples including Western samples and in different industries. Future research could examine both cyberbullying and FTF bullying perpetration as outcomes of POP and negative emotions and to consider which emotions might be more (or less) strongly related to each type of workplace bullying. Lastly, we relied upon self-reports for measuring the study variables (e.g., WCB perpetration), which might lead to common method bias. It has been argued that self- and other-ratings of antisocial work behavior generally result in a very similar pattern of findings (e.g., Berry et al., 2012). Moreover, we used a time-lagged research design, employed scales with good reliability and validity to measure the study variables, ensured respondents’ anonymity and confidentiality, and statistically controlled social desirability bias. These remedial measures made us confident that common method bias was not of particular concern for this study. Having said that, future studies may use multi-sourced data to indicate the frequency with which focal participants engage in cyberbullying behavior.


In conclusion, POP as a situational factor creates a stressful work environment that provokes anger and fear in employees and encourages both offline and online bullying in the workplace. Because both offline and online bulling are conceptually identical and co-occur in the workplace, thus an integrated approach might be more beneficial to prevent and reduce both forms of bullying at work. Further, intervention programs might be more effective and mutually reinforcing when they include both individual-directed as well as organizational-directed strategies.