Abstract
Existing workplace incivility research from the perspective of third parties has been found to be limited in explaining the specific boundary conditions of their constructive responses, such as punishing the instigator and helping the target. The lack of relevant knowledge hinders a comprehensive understanding of the coping strategies favoured by third parties. Building on the deontic model and the value protection model, this research involved two sub-studies with a total of 710 Chinese employees, employing a scenario experiment (Study 1) and a time-lagged survey (Study 2). The results showed that witnessed incivility positively predicted third parties’ workplace ostracism against the instigator, with moral anger acting as a mediator. The research did not identify a direct link between witnessed incivility and third parties’ organisational citizenship behaviour towards the target. However, the mediating role of moral anger between these two variables was found in Study 1. Moreover, Study 1 indicated that political skill strengthened the relationship between witnessed incivility and moral anger, but weakened the relationship between moral anger and third parties’ workplace ostracism against the instigator or their helping behaviour towards the target—findings partially supported by Study 2. These insights provide a practical and theoretical understanding of how organisations can utilise the role of third parties to intervene in workplace incivility effectively.
Similar content being viewed by others
Introduction
Workplace incivility refers to low-intensity deviant behaviours that violate social norms for mutual respect but lack a clear intention to harm (Andersson and Pearson 1999, p. 457). Currently, workplace incivility is believed to be more prevalent than other forms of mistreatment (Dhanani et al. 2021), and, as such, it may have detrimental implications for both employees and organisations (e.g., Cortina et al. 2022; Han et al. 2022; Mehmood et al. 2024). Research into incivility has traditionally centred on the direct individuals involved, namely instigators and targets (i.e., victims of mistreatment), whereas the influence of such mistreatment on third parties who witness these incidents has received relatively less attention (Miner et al. 2018; Schilpzand et al. 2016). Researchers have highlighted the importance of investigating third-party intervention in witnessed mistreatment, such as incivility, for a more complete picture of the field (Hill et al. 2025; Ni et al. 2024). The present research resonates with this call by examining third parties’ constructive responses to witnessed incivility—in particular, advocating for the victim by seeking punishment for the instigator or offering social support to the target (e.g., Jensen and Raver 2021; O’Reilly and Aquino 2011). This perspective is more consistent with the innate human tendency to support victims rather than ignore the situation or side with instigators (O’Reilly and Aquino 2011). In addition, third parties are often numerous (Dhanani et al. 2021). Their timely responses to workplace mistreatment can effectively curb the rapid spread of harmful behaviours, thereby contributing to a healthy work environment (Arman 2020; Ni et al. 2024).
Despite existing studies providing evidence that third parties may react constructively to witnessed incivility (e.g., Jensen and Raver 2021; Lin and Loi 2021; Reich and Hershcovis 2015), several knowledge gaps remain in this field. First, Hershcovis et al. (2017) have called for an expansion of the range of third-party deontic reactions to witnessed incivility. For example, attention should be given to more specific means of punishment beyond committing the same mistreatment as the instigator based on the principle of an eye for an eye (Lin and Loi 2021). In this regard, ostracising the instigator becomes a tool available to third parties. Research has proposed workplace ostracism as a covert form of retribution against unpopular team members, which may go unnoticed by those being ostracised (O’Reilly 2013; Robinson et al. 2013). Recent findings have also indicated that employees, acting as third parties, proactively exclude peers who are rude to customers (Ni et al. 2024). Considering the prevailing maxim that “the customer is always right”, third-party employees might exhibit amplified moral responses when customers are victims. Thus, it is essential to examine whether third parties view workplace ostracism as a means of punishment when both the instigator and the victim of incivility are internal members of the organisation.
Second, research on the moral agency of third parties, drawing on the deontic model (Folger 2001) and the value protection model (Skitka 2002), suggests that individuals confronted with morally violating situations, such as workplace mistreatment, experience moral anger—a transient state characterised by interconnected emotions like hostility and contempt. This emotional response prompts them to act constructively as a demonstration of their moral character (O’Reilly and Aquino 2011; Priesemuth 2013). Although moral anger is often linked to punitive action against instigators, its mediating role between witnessed workplace mistreatment and assistance behaviour remains unclear. Importantly, while most research has predominantly centred on sympathy as a predictor of third-party prosocial behaviour towards the victim (e.g., Hershcovis and Bhatnagar 2017; Hill et al. 2025), some theoretical models (O’Reilly and Aquino 2011) and empirical evidence (e.g., Mitchell et al. 2015; Reich and Hershcovis 2015) still imply that moral anger may also motivate assistance to targets of mistreatment. Therefore, a more robust evidence base is required to elucidate the role of moral anger in prompting varying constructive responses from third parties.
Third, since the introduction of the deontic model to explain third-party intervention in witnessed workplace mistreatment, researchers have progressively explored the boundary conditions of deontic effects (e.g., Mitchell et al. 2015; Liu et al. 2021; O’Reilly et al. 2016). Emerging research underscores the necessity for a comprehensive examination of whether deontic forces function across the different stages of this process (Lin and Loi 2021; Priesemuth and Schminke 2019). To extend this line of inquiry, it is imperative for scholars to pinpoint the factors that heighten third parties’ sensitivity to situations of injustice (e.g., workplace incivility), thereby eliciting their deontic emotions (Lin and Loi 2021). Current research mostly delineates the boundary conditions that diminish third-party moral anger, such as third parties’ exclusion beliefs about the targeted coworker (Mitchell et al. 2015) and the dark triad (Liu et al. 2021). In exploring strategies to enhance third parties’ moral emotions, scholars have predominantly emphasised the role of moral identity (e.g., Lin and Loi 2021; O’Reilly et al. 2016; Skarlicki and Rupp 2010), yet have stopped short of identifying additional factors that influence the relationship between witnessed incivility and moral anger.
Last, another important focus of research on the boundary conditions of third parties’ moral responses is the examination of the factors that influence the coping strategies they ultimately adopt (i.e., how moral anger translates into their behavioural responses; Lin and Loi 2021; Priesemuth and Schminke 2019). Some research has shown that third parties are more likely to choose punishment for the instigator when intervening in workplace mistreatment (e.g., Reich and Hershcovis 2015; van Prooijen 2010). Conversely, other studies have suggested that victim assistance is often favoured as the primary strategy (e.g., Dhaliwal et al. 2021; Artavia-Mora et al. 2017). The current issue remains underexplored, with a limited number of studies endeavouring to offer initial understanding by investigating the moderating influence of moral identity (Lin and Loi 2021; Mitchell et al. 2015).
To fill these research gaps, we evaluated our proposed model (shown in Fig. 1) across two studies using a multi-method approach to robustly confirm the reliability of our findings. Grounded in the deontic model and the value protection model, we argue that moral anger mediates the relationship between witnessed incivility and third parties’ workplace ostracism against the instigator (i.e., instigator-oriented punishment) or individually directed organisational citizenship behaviour (OCB-I) towards the target (i.e., target-oriented assistance). To clarify the factors shaping the deontic reactions of third parties, we propose political skill as a boundary condition for these effects. Given its role in keenly perceiving cues in the environment (Frieder et al. 2019; Munyon et al. 2015), political skill can enhance perceptions of unfair situations (e.g., workplace incivility). Additionally, this trait enables individuals to flexibly choose intervention strategies based on practical needs, effectively addressing unforeseen incidents in the workplace (Frieder et al. 2019; Munyon et al. 2015). We thus consider the dual functions of political skill in different stages of third-party reactions (i.e., the former in the deontic emotional reaction and the latter in behavioural reaction; Lin and Loi 2021)—thereby simultaneously examining its moderating role in both stages.
The conceptual model.
Our work makes several primary contributions. First, our research extends the literature on witnessed workplace mistreatment by revealing that workplace ostracism can be a strategic tool for third parties to penalise those who instigate incivility. This finding supports perspectives that organisational members will distance themselves from deviants within the organisation (O’Reilly 2013; Robinson et al. 2013) and answers recent calls for actively exploring the potential connections between various forms of workplace mistreatment, rather than viewing them in isolation (Cortina et al. 2017; Ferris et al. 2017; Zhan and Yan 2021). Second, our research contributes to the understanding of how moral anger prompts third parties’ responses to workplace mistreatment, especially among Chinese employees. By examining how moral anger elicits distinct constructive responses from third parties towards mistreated individuals, we evaluate the explanatory power of deontic theory in the Eastern workplace context. Last, our research broadens the scope of conditions influencing third-party moral reactions to workplace mistreatment. Unlike the usual emphasis on moral-related variables (e.g., Lin and Loi 2021; Mitchell et al. 2015), we investigate the role of political skill, an individual variable closely associated with organisational ecology, across different stages of the deontic process. This not only identifies sources of influence that can facilitate third-party intervention against incivility but also contributes to this ongoing debate surrounding “punishment or assistance primacy”.
Theoretical framework
The deontic model
Research is increasingly focusing on the role of third parties in workplace mistreatment, with the deontic model providing a key framework (Hill et al. 2025). This model conceptualises appropriate treatment as a moral imperative, inherently viewed as “the right thing to do” (Folger and Stein 2017). Therefore, witnessing workplace mistreatment elicits a sense of moral anger that can motivate retaliation against the instigator (Dhanani and LaPalme 2019; Hill et al. 2025). This concurs with research indicating that observed incivility is related to increased anger and antisocial behaviours directed at the instigator (e.g., Reich and Hershcovis 2015; Lin and Loi 2021; Ni et al. 2024). It is important to emphasise that, although the deontic theory is noted for its perpetrator-focused and anger-focused propositions (e.g., Hill et al. 2025), some researchers have argued that moral anger triggered by witnessed workplace mistreatment can also motivate individuals to engage in victim-focused assistance (e.g., Mitchell et al. 2015; Reich and Hershcovis 2015). This is consistent with the notion that both retributive and supportive responses are morally grounded and contribute to the preservation of the moral order (O’Reilly 2013; O’Reilly and Aquino 2011).
The value protection model
The value protection model (VPM) extends the deontic model’s understanding of third parties’ constructive responses in workplace mistreatment. The VPM highlights the importance of personal identity, particularly the need to maintain a positive personal identity (Skitka 2002). Given that moral values are integral to personal identity, observing workplace mistreatment, including incivility, poses a threat to the observer’s personal identity (Priesemuth 2013). Consequently, third parties may experience moral anger and resort to moral cleansing behaviours, such as penalising the instigator and offering support to the victim, in an effort to strengthen their beliefs in their moral integrity (i.e., their sense of being a good person; Skitka 2002; Skitka and Mullen 2002). Therefore, the VPM supports the deontic model’s premise that observers are driven by moral anger to make deontic responses to the witnessed mistreatment.
In addition to maintaining private personal identity, the VPM suggests that third-party constructive responses to workplace mistreatment serve another important function: contributing to the preservation of public personal identity by demonstrating moral stances to others (Semmer et al. 2019; Skitka 2002). This underscores the potential instrumental benefits of such actions, such as receiving gratitude from peers and thus gaining an improved reputation (Dhaliwal et al. 2021; Jordan and Rand 2020). These benefits help mitigate the limitations of the deontic model—which primarily emphasises purely altruistic motivations behind third parties’ constructive intervention. The VPM acknowledges that individuals may naturally engage in a “cost-benefit mental calculus” once they are aware that a situation might necessitate their involvement (Jensen and Raver 2021, p.886). This analysis can inform the selection of intervention strategies (Skarlicki and Kulik 2005). Consequently, the VPM provides a complementary perspective to the deontic model by explaining how third parties might strategically select intervention methods to maximise potential benefits, including improving their public image and reputation.
Literature review and hypotheses development
Witnessed workplace incivility and third parties’ workplace ostracism against the instigator
Workplace ostracism is formally defined as “the extent to which an individual perceives that he or she is ignored or excluded by others” (Ferris et al. 2008, p.1348). Previous studies have shown that third parties often rectify witnessed incivility by seeking punishment for instigators (O’Reilly and Aquino 2011; Reich et al. 2021; Reich and Hershcovis 2015). Inspired by these results, our research provides evidence that workplace ostracism may serve as a form of instigator-oriented punishment.
Evidence supporting this hypothesis can be found in Robinson et al. (2013) model, which demonstrates how organisational members employ “the silent treatment” as a well-known method of “purposeful ostracism” directed at deviant individuals to protect themselves and their teams. Similarly, O’Reilly (2013) proposes that third parties resort to “covert social sanctions” to impose a social cost on instigators of workplace mistreatment. Workplace ostracism, a form of such social sanctioning, is largely effective because it precludes instigators from capitalising on the benefits derived from their social networks (O’Reilly 2013). For example, third parties are hesitant to engage in future collaborations with individuals who exhibit incivility (Reich and Hershcovis 2015), thereby eroding the instigator’s reputation and social relationships, which in turn lead to an unintended social cost (Guo et al. 2021; O’Reilly 2013).
In spite of the widespread belief that third parties would ostracise the instigator of incivility (e.g., Ferris et al. 2017; Zhan and Yan 2021), empirical evidence in this area remains limited. Recent research indicates that third parties ostracise those who instigate incivility against customers (Ni et al. 2024), but this reaction appears to be influenced by the role of the victim as a customer. In this research, we examine whether third parties use such punishment in response to incivility between coworkers, a more prevalent and subtle violation of workplace interpersonal norms (Lin and Loi 2021; Reich and Hershcovis 2015). Thus, we hypothesise as follows:
Hypothesis 1. Witnessed workplace incivility is positively related to workplace ostracism against the instigator.
Witnessed workplace incivility and third parties’ individually directed organisational citizenship behaviour towards the target
Third parties’ punishment against instigators has garnered substantial scholarly interest, but the potential of another constructive intervention—helping the target—cannot be ignored. It is worth recognising that both punishing the wrongdoer and aiding the victim are considered constructive responses, as they can restore a sense of justice threatened by moral violations stemming from witnessed workplace mistreatment (O’Reilly 2013; O’Reilly and Aquino 2011). Furthermore, both forms of intervention can serve as means of “moral cleansing” that help third parties maintain their self-identity (Priesemuth 2013).
Previous research has provided evidence supporting this hypothesis. For instance, when third parties observe abusive supervision (Arman 2020; Priesemuth 2013), workplace incivility (Chui and Dietz 2014; Jensen and Raver 2021; Reich et al. 2021), workplace bullying (Coyne et al. 2019), or other forms of workplace mistreatment, they exhibit varying degrees of willingness to help the victim. Even with this abundant evidence, it is important to note that some studies have not clearly specified the manifestation of helping behaviours, leading to discrepancies in measurement (e.g., Chui and Dietz 2014). In this research, we employ the concept of “individually directed organisational citizenship behaviour (OCB-I)” introduced by Aryee et al. (2002) to assess the extent to which third parties offer assistance to targets of workplace incivility. This behaviour involves individuals actively providing aid without being constrained by organisational role requirements, thereby demonstrating their moral motivations (Bowes-Sperry and O’Leary-Kelly 2005). Given the above arguments, we propose that:
Hypothesis 2. Witnessed workplace incivility is positively related to OCB-I towards the target.
The mediating role of moral anger
The existing decision-making process models for third-party intervention in workplace mistreatment are mostly cognitive-oriented, assuming that third parties will decide on the intervention behaviours following a comprehensive evaluation of the incident (e.g., Bowes-Sperry and O’Leary-Kelly 2005; Skarlicki and Kulik 2005). The deontic model presents an alternative view, stating that the decision-making process for third-party intervention is not always rational or controllable. More often, it is merely a rapid instinctive response to unfair events driven by emotion, rather than a careful consideration of the benefits and risks associated with their behaviours (O’Reilly and Aquino 2011). Therefore, our research’s initial concern is examining the emotion-driven pathway of third-party intervention in workplace incivility, suggesting that the emotion aroused by unfair events can facilitate the initial motivation for individual intervention.
According to the deontic model, when witnessing workplace mistreatment, third parties are inclined to develop implicit moral intuitions that perceive the ethical impropriety and violations of such treatment. Simultaneously, such intuitions can elicit moral anger, a negative affective state encompassing various discrete emotions such as anger, hostility, and contempt (Hill et al. 2025). Consequently, this emotional response may spur observers to stop the mistreatment (O’Reilly and Aquino 2011). The VPM also presents similar opinions, suggesting that moral anger motivates third parties to respond to workplace mistreatment to protect personal moral values (Priesemuth 2013). Notably, although previous research has established a strong association between anger and punitive behaviours (Hill et al. 2025; Jordan and Rand 2020; Reich and Hershcovis 2015), some researchers have also highlighted that moral anger can drive individuals to assist victims (Mitchell et al. 2015), because both responses can be viewed as deontic reactions that play crucial roles in relieving third parties’ anger and facilitating their own sense of justice restoration (O’Reilly 2013; O’Reilly and Aquino 2011). Thus, we propose the following hypotheses:
Hypothesis 3a. Moral anger mediates the relationship between witnessed workplace incivility and workplace ostracism against the instigator.
Hypothesis 3b. Moral anger mediates the relationship between witnessed workplace incivility and OCB-I towards the target.
Intervention preference in workplace incivility: moral and self-interest motivations
While both punishment and assistance can restore a sense of justice after fairness norms have been violated, individuals may prefer different coping strategies: Some scholars suggest that punishing instigators is the most direct and effective way to make them face consequences (van Prooijen 2010). Moral anger is associated with instigator-focused attention, which often leads to the neglect of victims (Gummerum et al. 2016; Hill et al. 2025; Reich and Hershcovis 2015). This punitive tendency is also socially grounded: Instigators are generally considered threatening to society, and it aligns with third parties’ instinctive desire to punish them (van Prooijen 2010). However, the potential risks of harsh punishment may render these actions counterproductive. If third parties face the risk of retaliation (Artavia-Mora et al. 2017), or struggle with assigning clear responsibility for the observed mistreatment (Skarlicki and Kulik 2005), blindly implementing punishment may result in unfavourable outcomes. Additionally, helping behaviour, in contrast to punitive action that may deviate from moral values themselves (Lin and Loi 2021), better demonstrates altruism and becomes the superior strategy for third-party intervention (Dhaliwal et al. 2021; Jordan and Rand 2020).
The debate points out the limitations of the intuitive perspective: while moral anger can explain individuals’ initial moral motivations for intervention (O’Reilly and Aquino 2011), it falls short in explaining the variations in intervention methods among individuals. In fact, the deontic model does not exclude the instrumental motivations of third parties, and the decision-making process of intervention incorporates rational processing (Dhanani and LaPalme 2019; Folger and Stein 2017). When individuals strive to identify the most optimal constructive response, they often conduct cost-benefit analyses (Bowes-Sperry and O’Leary-Kelly 2005; Skarlicki and Kulik 2005). Recent research by Jensen and Raver (2021) challenges the notion of a complete separation of moral motivations and self-interest motivations in individual intervention against workplace incivility. Based on these findings, it is suggested that preference differences in third-party intervention strategies arise from the combined influence of both motives.
Self-interest motivations in third-party intervention: a VPM-based perspective
Third-party intervention in workplace mistreatment involves both an intrapsychic process, which concerns moral motives and moral anger arising from unfair incidents (i.e., the deontic emotional reaction), and an interpersonal behaviour process, where third parties consciously or unconsciously weigh the costs and benefits to choose the most effective strategy to exert influence through these actions (i.e., the behavioural reaction; Bowes-Sperry and O’Leary-Kelly 2005; Lin and Loi 2021; O’Reilly and Aquino 2011; Skarlicki and Kulik 2005). Building on the VPM and extending the perspective of the deontic model, we propose that third parties are instinctively driven to make constructive responses to workplace mistreatment in order to uphold moral principles. Such actions not only reinforce their positive self-evaluations but also communicate the moral stances of these individuals to others, leading to positive external evaluations and ultimately enhancing the reputation of these actors (Skitka 2002; Skitka and Mullen 2002). Even though self-interest motives might not constitute the initial impetus of third parties’ constructive responses, individuals’ morally driven actions can contribute to personal benefit, irrespective of their awareness (Jensen and Raver 2021; Skarlicki and Kulik 2005).
The moderating role of political skill
This research considers the functions of political skill in different stages of third-party reactions (i.e., the former in the intrapsychic process and the latter in the interpersonal behaviour process) and simultaneously examines its moderating role in both stages. Individuals with high political skill possess social astuteness, enabling them to comprehend others’ motives and desires. They also demonstrate interpersonal influence by adapting to and engaging in situationally appropriate behaviours, manifest sincerity through goodwill and trustworthiness, and establish beneficial alliances with key individuals using their networking abilities (Ferris et al. 2007). The influence of political skill extends beyond individuals to affect those around them as well. In other words, political skill is thought to shape not only individuals’ self-assessment and self-image but also the perceptions others have of them (Munyon et al. 2015; Frieder et al. 2019). This dual capacity aligns with individuals’ motivation to uphold moral values when intervening in incivility. Hence, political skill affects deontic responses of third parties through its moderating effects on both their emotional reactions to witnessed incivility and the relationship between emotional reactions and intervention behaviours.
Intrapsychic process: the influence of political skill on third parties’ emotional reactions
Political skill operates beginning with an individual’s self-assessment and situational assessment in specific contexts, collectively reflecting the intrapsychic effects of political skill (Munyon et al. 2015; Frieder et al. 2019). Those who possess high political skill are adept at interpreting and navigating the subtleties and complexities of personal motivation and the workplace environment (Ferris et al. 2007), a proficiency essential for evaluating the inherent ambiguity of workplace incivility.
Specifically, political skill influences an individual’s self-assessment of personal resources (Munyon et al. 2015; Frieder et al. 2019), with those exhibiting high political skill demonstrating an ability to comprehend and influence various aspects and elements of the work environment, thereby fostering a high level of self-efficacy (Munyon et al. 2015). This sense of control within the work environment empowers them to confidently address unforeseen challenges (e.g., witnessed incivility) and sustain the smooth operation of the organisation (Ferris et al. 2007). Moreover, third parties who exhibit stronger motivation for intervention against observed injustice are more likely to report increased levels of moral anger (Jordan and Rand 2020).
Political skill also plays a prominent role in impacting individuals’ evaluation and comprehension of their environment, particularly within the context of interpersonal relationships. Through perspective-taking, political skill empowers individuals to gain a deeper understanding of others’ preferences and needs (Frieder et al. 2019; Munyon et al. 2015). Research has identified a strong relationship between political skill and deep acting, which refers to the ability to regulate emotional expressions in accordance with situational demands while maintaining genuine emotion (Toomey et al. 2021). Therefore, it is postulated that individuals with higher political skill are more likely to exhibit emotions and attitudes congruent with the circumstances they encounter, especially in the face of incivility: They intuitively experience anger towards the instigator and stand in solidarity with the victims (Reich and Hershcovis 2015). This victim-oriented approach elicits a stronger perception of moral transgression, leading to increased levels of moral anger (Fiori et al. 2016; Reich et al. 2021). Based on the above evidence, we predict that:
Hypothesis 4a. Political skill moderates the relationship between witnessed incivility and moral anger, such that the relationship is weaker when political skill is lower rather than higher.
Interpersonal behaviour processes: the influence of political skill on third parties’ behavioural reactions
Following the intrapsychic effects, political skill further influences individuals’ interpersonal behaviour, primarily through prompting them to engage in behaviours that project a positive self-image to influence others’ evaluations (Frieder et al. 2019; Munyon et al. 2015). Individuals with high political skill exhibit a comprehensive comprehension of social interaction rules. This allows them to flexibly adjust their behaviour in diverse situations and with diverse individuals, thereby effectively employing influencing strategies (Ahearn et al. 2004; Ferris et al. 2007; McAllister et al. 2018).
Highly politically skilled individuals place a priority on maintaining interpersonal relationships (Wang and Luan 2024) and are less likely to confront instigators directly (Liu et al. 2008; Wu and Yang 2019). A workplace survey conducted in China revealed that employees with high political skill are tactful in handling matters and strive to maintain professional friendships with a wide range of colleagues (Liu et al. 2008). Even when confronted with witnessed incivility that incites moral anger, these employees refrain from using impulsive and destructive tactics, such as workplace ostracism, which may escalate conflicts and harm their reputation—a concept referred to as “face” in Chinese culture. Conversely, political skill predicts OCBs, enabling third parties to maintain high levels of helpfulness despite unpleasant organisational experiences (Kaur and Kang 2023; Moon and Morais 2022). By engaging in acts of assistance, individuals not only restore their own sense of justice but also cultivate a positive impression among their colleagues, a process that can be seen as a form of impression management (Brouer et al. 2015; Frieder et al. 2019; McAllister et al. 2018), leading to favourable evaluations. In conclusion, when observing workplace incivility, highly politically skilled individuals are less likely to resort to punitive behaviours rooted in moral anger. Instead, they tend to shift towards offering substantial assistance, safeguarding their own self-value, and sustaining positive social connections. Hence, we predict that:
Hypothesis 4b. Political skill moderates the relationship between moral anger and workplace ostracism against the instigator, such that the relationship is weaker when political skill is higher rather than lower.
Hypothesis 4c. Political skill moderates the relationship between moral anger and OCB-I towards the target, such that the relationship is weaker when political skill is lower rather than higher.
Together, Hypotheses 1-4 suggest a moderated mediation model. The indirect effect of witnessed workplace incivility on OCB-I towards the target through moral anger is lower for third parties with lower political skill. Regarding the indirect relationship between witnessed incivility and third parties’ ostracism against the instigator through moral anger, third parties’ political skill may amplify the first-stage relationship and alleviate the second-stage effect. Therefore, the relative magnitude of the two competing forces will determine the direction of the influence of political skill on this indirect relationship.
Hypothesis 5a (Exploratory). Political skill moderates the positive indirect effect of witnessed workplace incivility on workplace ostracism against the instigator through moral anger.
Hypothesis 5b. Political skill moderates the positive indirect effect of witnessed workplace incivility on OCB-I towards the target through moral anger, such that the indirect effect is weaker when political skill is lower rather than higher.
Study 1
Method
Participants and procedures
A single-factor two-level (condition: civil vs. uncivil) between-subjects design was employed in Study 1. We applied a scenario-simulation approach to manipulate the independent variable (witnessed incivility)—a technique validated in previous research (e.g., Lin and Loi 2021; Reich and Hershcovis 2015; Reich et al. 2021). We collected examples of civil and uncivil behaviours from established scales (e.g., Chen et al. 2019) and adapted them to the Chinese workplace culture to develop written vignettes depicting civil or uncivil workplace interactions (see Supplementary Appendix A). The scenarios were crafted without specifying the characters’ attributes, such as gender and position, to avoid influencing third-party interpretations (Lin and Loi 2021; Reich and Hershcovis 2015). Participants were randomly assigned to one of the two experimental conditions, and instructed to imagine witnessing the described interaction. They then evaluated their emotional and behavioural reactions and completed the relevant scales. When deciding the anticipated effect size, we referred to Reich and Hershcovis’ (2015) experiment, in which the effect sizes of incivility conditions were small for both punishment against the instigator (η2 = 0.08) and assistance towards the target (η2 = 0.05). Given these considerations, we conducted a priori power analysis in G*Power (Faul et al. 2009), specifying t-tests (difference between two independent means), a small effect size (d = 0.25), and an α of 0.05, which indicated that a minimum of 506 participants was needed to achieve 80% power.
Full-time employees were recruited from a state-owned company in Central China, with the support of the company’s management who encouraged employees to participate. Prior to the survey, we guaranteed participants the confidentiality and voluntary nature of our experiment. Our initial dataset included 562 participants, but we had to exclude 18 who failed to pass an attention check and 6 who had excessively short response times (i.e., <2 min). This left us with a final sample of 538 employees, of whom 60.59% were male. The age distribution was as follows: 31.97% were aged 18–29, 27.14% were aged 30–39, and 40.89% were aged 40 or above. Regarding educational background, 56.13% had high school diplomas or below, 37.17% held bachelor’s degrees, and 6.70% had master’s degrees or higher. In terms of tenure, 31.04% had less than 3 years, 13.75% had 3–5 years, and 55.21% had more than 5 years. The distribution of professional titles was as follows: 71.00% of the participants were at the junior level, 25.28% were at the intermediate level, and 3.72% were at the senior level.
Measures
Unless otherwise noted, items were rated on a 5-point Likert scale (1 = strongly disagree, 5 = strongly agree). The measures were translated into participants’ native language using the recommended back-translation procedures (Brislin 1970).
Manipulation check
In order to validate the manipulation, all participants were asked to rate the degree of incivility of each episode on a 5-point Likert scale (1 = no incivility, 5 = extreme incivility).
Moral anger
We used a three-item scale adapted from Raver et al. (2012) to assess participants’ moral anger towards the scenario. An example item is “I feel angry about the scenario.” Cronbach’s α was 0.887. The individual measurement model for moral anger was a saturated model: CFI = 1.000; IFI = 1.000; SRMR = 0.000; RMSEA = 0.000.
Workplace ostracism against the instigator
We adapted Jiang et al. (2011) nine-item scale to measure participants’ proactive exclusion against the instigator within the scenario. This scale is based on the Workplace Ostracism Scale developed by Ferris et al. (2008) and has been adapted to Chinese cultural contexts. An example item is “I avoid the instigator at work.” Cronbach’s α was 0.928. The initial individual measurement model fit was χ2/df = 16.211; CFI = 0.880; IFI = 0.880; SRMR = 0.059; RMSEA = 0.168. Model fit improvement was conducted by correlating error terms based on the modification indices (Collier 2020). All modifications were made in accordance with realistic and theoretically justifiable interpretations. The revised individual measurement model for workplace ostracism against the instigator yielded a good fit for the data: χ2/df = 5.075; CFI = 0.973; IFI = 0.973; SRMR = 0.033; RMSEA = 0.087.
OCB-I towards the target
We used a 4-item scale pooled from the extant literature (e.g., Aryee et al. 2002, 2024; Henderson et al. 2020) to measure participants’ assistance towards the target within the scenario. An example item is “I voluntarily help the target settle into the job.” Cronbach’s α was 0.897. The individual measurement model fit for OCB-I was χ2/df = 11.565; CFI = 0.984; IFI = 0.984; SRMR = 0.020; RMSEA = 0.140. The suboptimal performance of the RMSEA may be ascribed to the small degrees of freedom (Kenny et al. 2015).
Political skill
Participants were asked to indicate their agreement with a six-item measure of political skill (Ahearn et al. 2004). An example item is “I find it easy to envision myself in the position of others.” Cronbach’s α was 0.903. The initial individual measurement model fit was χ2/df = 6.722; CFI = 0.972; IFI = 0.972; SRMR = 0.030; RMSEA = 0.104. Model fit improvement was conducted by correlating error terms based on the modification indices, and the CFA confirmed that the revised unidimensional structure is a good fit for the data: χ2/df = 3.617; CFI = 0.989; IFI = 0.989; SRMR = 0.020; RMSEA = 0.070.
Control variables
Third parties’ socio-demographic variables have been shown to influence their responses to the witnessed workplace mistreatment (Chui and Dietz 2014; Reich and Hershcovis 2015). Therefore, participants’ gender, age, education, tenure, and position were collected and controlled.
Results
Manipulation check
An independent samples t-test was conducted to evaluate the effectiveness of the incivility manipulation by using incivility condition (uncivil or civil) as the independent variable and participants’ incivility rating of the scenario as the dependent variable. Participants in the uncivil condition (M = 4.74, SD = 0.51) perceived that the instigator behaved more uncivilly towards the target than those in the civil condition (M = 1.36, SD = 0.56), t(491.18) = −72.22, p < 0.001. Hence, we concluded that participants correctly perceived the incivility condition.
Measurement model evaluation
We evaluated construct reliability using composite reliability (CR) and average variance extracted (AVE). Table 1 shows that the CR and AVE for all constructs exceed the recommended thresholds of 0.7 and 0.5, respectively. The factor loadings of the items exceeded 0.5 (Cortina et al. 2001; Ferris et al. 2005). To evaluate the discriminant validity of the study variables (moral anger, workplace ostracism against the instigator, OCB-I, and political skill), we conducted a confirmatory factor analysis (CFA) using AMOS. A four-factor measurement model provided satisfactory model fit (χ2/df = 4.038; CFI = 0.926; IFI = 0.926; SRMR = 0.044; RMSEA = 0.075). The model fit of the four-factor model was better than that of all the other alternative measurement models, including the three-factor model combining workplace ostracism and OCB-I (χ2/df = 11.430; CFI = 0.741; IFI = 0.741; SRMR = 0.173; RMSEA = 0.139) and the two-factor model combining workplace ostracism, OCB-I and political skill (χ2/df = 19.444; CFI = 0.537; IFI = 0.538; SRMR = 0.216; RMSEA = 0.185). The one-factor model yielded a poor fit (χ2/df = 21.962; CFI = 0.471; IFI = 0.472; SRMR = 0.222; RMSEA = 0.198). Therefore, the study variables are conceptually distinctive.
Hypothesis testing
Table 2 presents the means, standard deviations, and intercorrelations for the variables in Study 1. We employed the PROCESS macro for SPSS (Hayes 2018) to examine conditional indirect effects within moderated mediation models and assessed our hypothesis. We used 5000 bootstrapped samples to calculate the point estimates and bias-corrected confidence intervals (CIs) for the indirect effects.
Mediating effect analysis
We used Model 4 to examine simple mediation. As shown in Table 3, witnessed incivility had a significant positive effect on workplace ostracism against the instigator (β = 1.18, p < 0.001, 95% CI [1.01, 1.34]), supporting Hypothesis 1. Consistent with Hypothesis 3a, moral anger mediated the effect of witnessed incivility on workplace ostracism against the instigator (indirect effect = 0.58, Boot SE = 0.07, 95% CI [0.44, 0.72]). Contrary to Hypothesis 2, witnessed incivility had a significant negative effect on OCB-I towards the target (β = −0.55, p < 0.001, 95% CI [−0.75, −0.36]). Moral anger still mediated the relationship between these two variables (indirect effect = 0.24, Boot SE = 0.09, 95% CI [0.05, 0.40]), thus supporting Hypothesis 3b.
Moderated mediating effect analysis
We employed Model 58 to verify the moderating role of political skill and the validity of the whole moderated mediation model. Consistent with Hypothesis 4a, the interaction between witnessed incivility and political skill on moral anger was significant (β = 0.57, p < 0.001, 95% CI [0.40, 0.74]). As shown in Fig. 2, witnessed incivility had a stronger positive effect on moral anger at higher levels of political skill (simple slope = 1.83, p < 0.001) than at lower levels (simple slope = 0.70, p < 0.001). Consistent with Hypothesis 4b, the interaction between moral anger and political skill on workplace ostracism against the instigator was significant (β = −0.11, p < 0.001, 95% CI [−0.20, −0.03]). The simple slope analysis revealed that moral anger had a stronger positive effect on workplace ostracism against the instigator at lower levels of political skill (simple slope = 0.57, p < 0.001), and the effect diminished when political skill was higher (simple slope = 0.34, p < 0.001). Hypothesis 4c was not supported, as the interaction between moral anger and political skill had a negative effect on OCB-I towards the target (β = −0.18, p < 0.001, 95% CI [−0.27, −0.08]). Moral anger had a stronger positive effect on OCB-I towards the target at lower levels of political skill (simple slope = 0.25, p < 0.001) compared to higher levels (simple slope = −0.11, p < 0.05).
a Moderating effect of political skill on the relationship between witnessed incivility and moral anger. b Moderating effect of political skill on the relationship between moral anger and workplace ostracism against the instigator. c Moderating effect of political skill on the relationship between moral anger and OCB-I towards the target. High/low = 1 SD above/below the mean.
We then compared the indirect effects across different levels of political skill. Moral anger played a mediating role between witnessed incivility and workplace ostracism against the instigator at both lower levels (indirect effect = 0.40, Boot SE = 0.08, 95% CI [0.25, 0.56]) and higher levels (indirect effect = 0.62, Boot SE = 0.13, 95% CI [0.37, 0.89]) of political skill. The difference between the two effects was not statistically significant (Diff = 0.22, Boot SE = 0.14, 95% CI [−0.05, 0.24]), failing to support Hypothesis 5a. In addition, moral anger played a mediating role between witnessed incivility and OCB-I towards the target at lower levels (indirect effect = 0.18, Boot SE = 0.06, 95% CI [0.06, 0.31]) instead of higher levels (indirect effect = −0.21, Boot SE = 0.14, 95% CI [−0.48, 0.08]) of political skill, leading to a statistically significant difference between these two effects (Diff = −0.39, Boot SE = 0.15, 95% CI [−0.67, −0.08]), in support of Hypothesis 5b. Figures 3 and 4 provide a visual representation of the conditional indirect, direct and total effects across different moderator levels.
Witnessed incivility and workplace ostracism against the instigator: direct, indirect (via moral anger), and total effects moderated by political skill.
Witnessed incivility and OCB-I towards the target: direct, indirect (via moral anger), and total effects moderated by political skill.
Study 1 discussion
Study 1 has revealed that moral anger plays a mediating role between witnessed workplace incivility and the subsequent coping strategies, with political skill acting as a moderator in this relationship. However, there are some limitations in Study 1 that prompt us to conduct a second study. First, the evaluation in Study 1 of participants’ responses to fictional uncivil scenarios may suffer from limited ecological validity. Consequently, Study 2 asks participants to recall their actual experiences dealing with real-life uncivil incidents. Second, considering the research trends that explore political skill at the dimensional level (Ferris et al. 2019; Frieder et al. 2019; McAllister et al. 2018), Study 2 aims to explore the moderating effects of the four sub-dimensions of political skill: social astuteness, interpersonal influence, networking ability, and apparent sincerity (Ferris et al. 2005). Despite the lack of existing theoretical knowledge on the distinct mechanisms of these dimensions, conducting exploratory or post-hoc data analyses to replace the omnibus political skill construct with its four specific dimensions in empirical research could enhance understanding in this area (Frieder et al. 2019, p. 559).
Study 2
Sample and procedure
Data were collected from full-time employees of a company in Western China through snowball sampling, with participants completing two online surveys, 1 week apart. This convenience sampling technique has been successfully utilised by researchers studying workplace mistreatment (e.g., Mitchell et al. 2015). Prior to the Time 1 survey, a critical incident technique was employed to ask participants to recall instances of uncivil interactions between two colleagues that they had witnessed at the workplace, involving impolite language, unfriendly attitudes, and offensive gestures, with explicit instructions to exclude intended aggression or physical assaults. Specifically, this technique required participants to recall details (e.g., time, venue, and process) of the witnessed incivility to enhance retrospective accuracy and vividness of their reports (Lin and Loi 2021; Mitchell et al. 2015). Participants provided the initials of the instigator and target in that situation, which were automatically inserted into the Time 1 and Time 2 survey items. For instance, when assessing ostracism towards the instigator, respondents were asked to refer to the situation and individuals identified in the situation they had provided. In addition, the critical incident description provided by the respondents at the beginning of the Time 1 survey was presented to them again at the start of the Time 2 survey for reference. Participants who reported not having directly witnessed incivility were excluded from the analysis.
In the first survey, participants completed measures of witnessed incivility, moral anger, political skill, and demographics. The second survey collected data on third-party responses to the mistreatment, including workplace ostracism against the instigator and OCB-I towards the target. Researchers recommend the use of time separation between predictor and criterion variables to reduce common method variance concerns (Podsakoff et al. 2012). A 1-week separation was employed because shorter intervals are believed to enhance the accuracy of participants’ self-reported information regarding low-intensity behaviours (Meier and Spector 2013).
At Time 1, we initially collected 193 responses, from which 4 were excluded due to insufficient incivility detail, and 2 failed the attention check, yielding 187 valid responses. At Time 2, we received 172 matched responses. Among the participants, 51.16% were male. The age distribution was as follows: 51.74% were aged 18–29, 11.05% were 30–39, 37.21% were aged 40 or above. Regarding educational background, 32.56% had high school diplomas or below, 54.07% held bachelor’s degrees, and 13.37% had master’s degrees or higher. In terms of tenure, 51.16% had less than 3 years, 1.75% had 3–5 years, 47.09% had more than 5 years. The distribution of professional titles was as follows: 72.67% of the participants were at the junior level, 26.16% were at the intermediate level, and 1.17% were at the senior level.
Measures
Unless otherwise noted, items were rated on a 5-point Likert scale (1 = strongly disagree, 5 = strongly agree).
Witnessed incivility (T1)
We used a 7-item scale adapted from previous research (Cortina et al. 2001). Third parties were asked to rate how often they were aware of the instigator (identified in the critical incident) engaging in each of the listed forms of incivility against the target (e.g., “Put the target down or was condescending to the target”). In Study 2, we deleted one item with low factor loading: “Make unwanted attempts to draw the target into a discussion of personal matters” (factor loading = 0.294). Cronbach’s α was 0.901. The initial individual measurement model fit was χ2/df = 7.787; CFI = 0.904; IFI = 0.905; SRMR = 0.059; RMSEA = 0.199. We adopted the same model refinement strategy as Study 1, confirming a good fit for the data with the revised unidimensional structure: χ²/df = 2.203; CFI = 0.987; IFI = 0.987; SRMR = 0.021; RMSEA = 0.084.
Moral anger (T1)
Same as Study 1. Cronbach’s α was 0.839. The individual measurement model for moral anger was a saturated model: CFI = 1.000; IFI = 1.000; SRMR = 0.000; RMSEA = 0.000.
Workplace ostracism against the instigator (T2)
Same as Study 1. In Study 2, we deleted one item with low factor loading: “When the instigator achieves success at work, I choose not to offer congratulations (factor loading = 0.458)”. Cronbach’s α was 0.892. The revised individual measurement model yielded a good fit for the data: χ2/df = 1.946; CFI = 0.985; IFI = 0.985; SRMR = 0.034; RMSEA = 0.074.
OCB-I towards the target (T2)
Same as Study 1. Cronbach’s α was 0.865. The CFA confirmed that the unidimensional structure is a good fit for the data: χ2/df = 1.222; CFI = 0.999; IFI = 0.999; SRMR = 0.016; RMSEA = 0.036.
Political skill (T1)
Participants were asked to indicate their agreement with an 18-item measure of political skill which defined the four dimensions of political skill: social astuteness (e.g., “I understand people very well”), interpersonal influence (e.g., “I am able to make most people feel comfortable and at ease around me”), networking ability (e.g., “I am good at building relationships with influential people at work”), and apparent sincerity (e.g., “I try to be genuine in what I say and do”) (Ferris et al. 2005). The Cronbach’s α for the total scale was 0.888, with the α values for each factor dimension being 0.767, 0.846, 0.835, and 0.725, respectively. The initial individual measurement model fit was χ2/df = 2.523; CFI = 0.847; IFI = 0.849; SRMR = 0.082; RMSEA = 0.094. The same model refinement strategy was employed, with the revised four-dimensional structure yielding a good fit for the data: χ2/df = 1.891; CFI = 0.912; IFI = 0.912; SRMR = 0.076; RMSEA = 0.072.
Control variables
Participants’ gender, age, education, tenure, and position were included as control variables.
Results
Common method bias check
Harman’s one-factor test was used to explore whether our results were susceptible to common method bias. Common method variance would be considered if an exploratory factor analysis (EFA) produced one factor that accounts for more than 40% of the covariance among the variables. In our case, all factors with eigenvalues greater than 1.0 collectively accounted for 70.34% of the variance, and the largest retained factor explained 18.73% of the total variance.
Measurement model evaluation
Table 4 shows that the CR and AVE for all constructs exceed the recommended thresholds of 0.7 and 0.5, respectively. The factor loadings of the items are higher than the threshold of 0.5 (Cortina et al. 2001; Ferris et al. 2005). Subsequently, a confirmatory factor analysis (CFA) was conducted to examine the construct validity of our five measures (witnessed workplace incivility, moral anger, political skill, workplace ostracism and OCB-I). We first loaded items onto their corresponding latent constructs, but the CFA model did not converge, possibly due to the small sample size. It is recommended that the parameter-to-sample size ratio meet or exceed the recommended ratio of 1:5 (Liu et al. 2019). To handle this issue, we parcelled the items of political skill based on dimensions using the internal-consistency approach (Kishton and Widaman 1994). The CFA results showed that the refined 5-factor model fit the data well (χ2/df = 1.889; CFI = 0.906; TLI = 0.907; SRMR = 0.077; RMSEA = 0.072). The model fit of the five-factor model was better than that of all the other alternative measurement models, including the four-factor model combining workplace ostracism and OCB-I (χ2/df = 3.806; CFI = 0.690; IFI = 0.694; SRMR = 0.116; RMSEA = 0.128) and the four-factor model combining moral anger and political skill (χ2/df = 2.990; CFI = 0.625; IFI = 0.629; SRMR = 0.101; RMSEA = 0.108). The one-factor model yielded a poor fit (χ2/df = 6.520; CFI = 0.377; IFI = 0.383; SRMR = 0.159; RMSEA = 0.180). Therefore, the study variables are conceptually distinct.
Hypothesis testing
Means, standard deviations, and intercorrelations of the variables in Study 2 are presented in Table 2.
Mediating effect analysis
We used Model 4 to examine simple mediation. As shown in Table 3, witnessed incivility had a significant positive effect on workplace ostracism against the instigator (β = 0.25, p < 0.001, 95% CI [0.11, 0.40]), supporting Hypothesis 1. Consistent with Hypothesis 3a, moral anger mediated the effect of witnessed incivility on workplace ostracism against the instigator (indirect effect = 0.06, Boot SE = 0.03, 95% CI [0.02, 0.13]). Neither witnessed incivility (β = 0.06, p = 0.44, 95% CI [−0.12, 0.25]) nor moral anger (β = 0.05, p = 0.60, 95% CI [−0.12, 0.23]) had a significant effect on OCB-I towards the target. Furthermore, the mediating role of moral anger in the relationship between witnessed incivility and OCB-I towards the target was non-significant (indirect effect = 0.01, Boot SE = 0.02, 95% CI [−0.04, 0.06]). Consequently, Hypotheses 2 and 3b were not supported. Given that our findings only supported the mediating role of moral anger between witnessed incivility and workplace ostracism against the instigator, we conducted a structural equation modelling (SEM) analysis on this mediating model and enhanced the model fit by correlating error terms as suggested by modification indices (Collier 2020). The results showed the revised model had good fit indicators (χ2/df = 1.979; CFI = 0.941; IFI = 0.942; SRMR = 0.065; RMSEA = 0.076).
Additionally, a post hoc power analysis was conducted using the tool developed by Schoemann et al. (2017) to assess the power in our mediation model. For this purpose, we chose values of 5000 and 20,000 for the total number of power analysis replications and the number of coefficient draws per replication respectively. Furthermore, standardised coefficients were incorporated into the model to ensure methodologically rigorous estimation. The post hoc power analysis revealed that with a sample size of 172, Study 2 achieved a power level of 0.83 at a two-tailed significance level of α = 0.05 for the indirect effect mediated by moral anger.
Moderated mediating effect analysis
We employed Model 58 to examine the moderating role of political skill and the validity of the whole moderated mediation model. Consistent with Hypothesis 4a, the interaction between witnessed incivility and political skill on moral anger was significant (β = 0.12, p < 0.05, 95% CI [0.01, 0.25]). The simple slope analysis revealed that witnessed incivility had a stronger positive effect on moral anger at higher levels of political skill (simple slope = 0.39, p < 0.001) compared to lower levels (simple slope = 0.15, p = 0.10) (see Fig. 5). The interactions between moral anger and political skill on workplace ostracism against the instigator (β = −0.02, p = 0.79, 95% CI [−0.13, 0.08]) and OCB-I towards the target (β = −0.12, p = 0.08, 95% CI [−0.26, −0.01]) were non-significant, providing limited support for Hypotheses 4b and 4c.
a Moderating effect of political skill on the relationship between witnessed incivility and moral anger. b Moderating effect of apparent sincerity on the relationship between witnessed incivility and moral anger. High/low = 1 SD above/below the mean.
We further examined the moderating effects of the four dimensions of political skill. Except for apparent sincerity (β = 0.24, p < 0.001, 95% CI [0.11, 0.38]), the interaction between witnessed incivility and the other three dimensions of political skill (social astuteness: β = 0.04, p = 0.57, 95% CI [−0.12, 0.19]; interpersonal influence: β = 0.13, p = 0.06, 95% CI [−0.01, 0.27]; networking ability: β = 0.09, p = 0.15, 95% CI [−0.04, 0.22]) on moral anger was non-significant, partially supporting Hypothesis 4a. As shown in Fig. 5, witnessed incivility had a stronger positive effect on moral anger under conditions of higher levels of apparent sincerity (simple slope = 0.54, p < 0.001) than under lower levels (simple slope = 0.05, p = 0.61). Hypothesis 4b was not supported because the interaction between moral anger and each dimension of political skill (social astuteness: β = −0.01, p = 0.81, 95% CI [−0.15, 0.07]; interpersonal influence: β = 0.01, p = 0.90, 95% CI [−0.12, 0.12]; networking ability: β = −0.03, p = 0.56, 95% CI [−0.16, 0.07]; apparent sincerity: β = 0.06, p = 0.30, 95% CI [−0.04, 0.17]) on workplace ostracism against the instigator was non-significant. In addition, the interaction between moral anger and each dimension of political skill (social astuteness: β = −0.04, p = 0.53, 95% CI [−0.20, 0.08]; interpersonal influence: β = −0.14, p = 0.07, 95% CI [−0.27, −0.01]; networking ability: β = −0.08, p = 0.23, 95% CI [−0.19, 0.03]; apparent sincerity: β = −0.12, p = 0.08, 95% CI [−0.24, 0.01]) on OCB-I towards the target was non-significant, failing to support Hypothesis 4c.
Considering the evidence for Hypotheses 3a and 4a, we further examined the validity of the whole moderated mediation model by using Model 7. To account for the moderating effect involving latent variables, interaction terms must be constructed by pairing product indicators in AMOS. However, the variety of strategies for generating these indicators can affect analytical outcomes. Thus, we utilised Mplus to conduct a latent moderated SEM. We developed both a baseline model and an enhanced model incorporating latent moderation terms, and then compared their fit indicators. The baseline model was found to fit well (χ2/df = 1.931; CFI = 0.919; TLI = 0.921; SRMR = 0.075; RMSEA = 0.074). The comparison of the Akaike information criterion (AIC) values between the two models revealed a difference of less than 2 (baseline: AIC = 10,627.003; enhanced: AIC = 10,628.980), suggesting no significant superiority of the baseline model over the enhanced model with latent moderation terms (Bevans 2023). Consequently, the SEM with latent moderation from Study 2 demonstrated a good fit.
The moderated mediation model test revealed that moral anger played a mediating role between witnessed incivility and workplace ostracism against the instigator at higher levels of political skill (indirect effect = 0.09, Boot SE = 0.04, 95% CI [0.02, 0.18]) compared to lower levels (indirect effect = 0.04, Boot SE = 0.03, 95% CI [−0.01, 0.10]), leading to a statistically significant difference between these two effects (Diff = 0.05, Boot SE = 0.04, 95% CI [0.01, 0.14]). When using apparent sincerity as a moderator, the same result pattern was observed: Moral anger mediated the relationship between witnessed incivility and workplace ostracism against the instigator. However, this mediation effect was stronger under conditions of higher levels of apparent sincerity (indirect effect = 0.13, Boot SE = 0.05, 95% CI [0.04, 0.23]) compared to lower levels (indirect effect = 0.01, Boot SE = 0.02, 95% CI [−0.04, 0.06]). The difference between these two effects was statistically significant (Diff = 0.12, Boot SE = 0.05, 95% CI [0.03, 0.23]). In conclusion, the validity of this moderated mediation model has been confirmed, partially supporting Hypothesis 5a. Since Hypotheses 3b, 4a and 4c were not supported, the moderated mediation model proposed in Hypothesis 5b was also not established.
Table 5 summarises the assessments across two studies. Detailed interpretations of discrepancies (e.g., Hypotheses 3b and 4b) are discussed in the following section.
General discussion
We conducted two studies to investigate the effects of moral anger and political skill on third parties’ constructive responses to workplace incivility, employing a multi-method approach. The results largely supported our hypotheses, providing valuable insights into the intricate dynamics of workplace mistreatment and underscoring the need to consider individual differences in interpersonal interaction style and social efficacy when examining such phenomena. Overall, our research holds important theoretical and practical implications.
Theoretical implications
Our research makes contributions to the literature in the following ways. First, we examine the relationship between workplace incivility and workplace ostracism through the lens of third parties, providing insights into linking these two research fields (Cortina et al. 2017; Zhan and Yan 2021). Specifically, we identify instigator-oriented workplace ostracism as a potential deontic reaction by third parties to witnessed incivility. Previous research in Western workplace cultures has suggested that ostracising deviant members may serve to protect organisational functioning (O’Reilly 2013; Robinson et al. 2013), yet this may not apply within the Chinese cultural context. On the one hand, Chinese employees, guided by Confucian etiquette norms, may demonstrate greater sensitivity to workplace incivility, leading to more severe negative assessments of instigators and a stronger tendency to retaliate (Lin and Loi 2021; Zhan and Yan 2021). On the other hand, the cultural emphasis on harmony often induces employees to ignore incivility to avert conflict (Ferris et al. 2017; Iqbal et al. 2023). Our findings support the expectation that Chinese employees, as third parties, are prone to experiencing moral anger and taking action against instigators following the observation of incivility (Study 1), and that these responses exhibit durability over time (Study 2). Furthermore, our findings complement Ni et al.’s (2024) study by showing that third-party ostracism of incivility instigators is consistent regardless of whether the target of incivility is an external organisational member (e.g., a customer) or an internal one (e.g., a co-worker).
Second, the findings from our two sub-studies provide limited evidence to suggest that third parties react positively towards targets. Notably, this is consistent with previous research in the field, which indicates that the direct impact of observed mistreatment on target-oriented assistance is not significant (e.g., Mitchell et al. 2015; Priesemuth and Schminke 2019; Reich and Hershcovis 2015). These results imply that observers may not view low-intensity and ambiguous mistreatment, such as workplace incivility, as severe enough to warrant special attention to victims. Prior studies suggest that third parties are more likely to support the target only when they receive ambiguity-reduction cues (e.g., the target reacts hurt; Coyne et al. 2019; Jensen and Raver 2021). Conversely, in instances of subtle workplace incivility, third parties might struggle to assign responsibility for the mistreatment, potentially attributing blame to the two parties involved, which can result in indifference towards the victim or even a bias in favour of the instigator (Liu et al. 2021; Ni et al. 2024). Moreover, research has shown that merely witnessing workplace mistreatment can have negative effects on individuals (Dhanani et al. 2021), such as evoking fear (Gabriel et al. 2024), thereby reducing the likelihood of observers engaging in OCBs (Cortina et al. 2017). In addition, Reich and Hershcovis (2015) remark that third-party responses to targets of incivility should demonstrate “symmetry” with the specific experiences of those subjected to such acts, indicating that observers are more likely to respond to these individuals with more civility rather than through other forms of compensation. In essence, our study contributes novel insights into the responses of third parties to witnessed workplace incivility.
Third, our research extends the literature on the deontic model by examining the role of moral anger in motivating third-party constructive intervention in witnessed incivility. Consistent with the most recent meta-analysis (Hill et al. 2025), we find across two sub-studies that moral anger reliably mediates the relationship between witnessed incivility and punishment against the instigator. These findings support the notion that witnessed mistreatment, as a violation of moral norms, elicits a heightened state of arousal in the observer (Gabriel et al. 2024; Hill et al. 2025), which is closely associated with a strong inclination to punish the instigator (e.g., Priesemuth and Schminke 2019). Drawing on the perspectives of several researchers (e.g., O’Reilly and Aquino 2011; Mitchell et al. 2015; Reich and Hershcovis 2015), we also examine the role of moral anger in linking witnessed incivility to another deontic response—OCB-I. The results of this examination are mixed: We find an indirect effect of witnessed incivility on target-oriented assistance through moral anger in Study 1, but not in Study 2. This corresponds with Mitchell et al.’s (2015) observation that the mediating role of anger in the link between observed mistreatment and support for coworkers is dependent on particular conditions, such as when the third party holds a strong moral identity. These findings also suggest that anger is more closely associated with punishment rather than assistance (e.g., Jordan and Rand 2020; Skarlicki and Kulik 2005; van Prooijen 2010). Additionally, unlike in Study 1, a temporal discrepancy between the measurement of third-party emotional states and their subsequent behaviours is intentionally introduced in Study 2, reflecting that moral anger may be more likely to prompt immediate rather than delayed assistance behaviours. This is consistent with the proposition that moral anger is a transient emotional surge that compels rapid behavioural responses (Hill et al. 2025).
Fourth, the results of our research also add to the expanding literature on the boundary conditions of various stages of deontic effects (e.g., Lin and Loi 2021; Mitchell et al. 2015; Priesemuth and Schminke 2019). In particular, we direct our attention to a previously overlooked moderating variable—political skill—and simultaneously examine its influence on the two key stages of third parties’ deontic reactions to witnessed workplace incivility: the emotional and behavioural responses (Lin and Loi 2021). Concerning the emotional response, two sub-studies consistently reveal that political skill intensifies the deontic emotions felt by third parties witnessing workplace incivility—specifically, a more intense moral anger. This finding is consistent with the meta-theoretical framework of political skill’s effects (e.g., Munyon et al. 2015; Frieder et al. 2019), indicating that this individual attribute can impact intrapsychic processes—fostering a deeper understanding of oneself, others, and the work environment. Possessing this trait enhances individuals’ ability to perceive ambiguous workplace incivility and more adeptly understand their emotions. As shown by Lin and Loi (2021) and further supported by Jeong et al. (2023), such heightened unfairness perception and negative emotional interpretation can lead to stronger deontic emotions, thereby motivating constructive actions to protect the private self-identity of third parties—a perspective that aligns with the essence of the deontic model and the value-protection model (Priesemuth 2013).
However, third parties inevitably engage in a cost-benefit analysis when translating deontic emotions into constructive actions (Jensen and Raver 2021). One potential aspect of this analysis is the selection of the most self-serving means to maintain one’s public personal identity, thereby presenting a positive image to a wider audience (Skitka 2002). Our two sub-studies preliminarily confirm that political skill influences individuals’ strategic preferences for constructive intervention in witnessed incivility. Specifically, the behaviour of highly politically skilled third parties is not solely driven by emotion, as they also prioritise saving face and maintaining harmonious relationships with others (Liu et al. 2008). Although workplace ostracism is a relatively subtle form of punishment (O’Reilly 2013), it may still expose third parties to the risk of escalating conflicts with the instigator and be deemed a form of destructive punitive reaction (Lin and Loi 2021). This coping strategy is detrimental to the construction of their workplace network (Su et al. 2021), and thus it is avoided by them (as confirmed in Study 1).
Contrary to the expected hypothesis, the indirect relationship between witnessed incivility and the third party’s OCB-I towards the target via moral anger is weakened when political skill is high versus low (as confirmed in Study 1, with Study 2 showing marginally significant results). This may be due to third parties’ concerns about potential conflict and retaliation from instigators, leading them to moderate their support for victims due to self-protection concerns (O’Reilly 2013). Nevertheless, in a supplementary analysis of Study 1, we found that individuals with higher political skill (top 27% in political skill scores) still exhibited higher levels of OCB-I than those with lower political skill (bottom 27%) after witnessing workplace incivility (t(160) = 6.858, p < 0.001). This is consistent with established literature on political skill, suggesting that highly politically skilled third parties consistently demonstrate high levels of OCB (Kaur and Kang, 2023; Moon and Morais 2022) in an effort to engage in impression management and establish a positive reputation within the organisation (Brouer et al. 2015; Frieder et al. 2019). Overall, we identify an intriguing pattern among individuals with high political skill in dealing with witnessed incivility: they adopt a nuanced approach—neither confronting the instigator directly nor offering excessive support to the victim, but rather seeking to maintain a balanced stance that preserves relationships with all parties involved, thus fostering a broader network of connections.
Last, we provide evidence supporting the moderating effect of one dimension of political skill—apparent sincerity. However, the conclusion should be interpreted with caution because although individuals with high apparent sincerity report higher levels of moral anger, their self-reported emotional states may not always align with their true feelings (Jordan and Rand 2020). Their reports of intense moral anger may be solely a form of impression management to portray themselves as individuals striving for justice, rather than an authentic expression of such emotion. Future research is needed to corroborate these hypotheses with additional evidence.
Practical implications
Our findings have significant implications for workplace practice. Whereas third parties’ intervention in workplace incivility can raise the costs of rude behaviours and serve as deterrents to instigators (Reich and Hershcovis 2015), inappropriate reactions from third parties can have detrimental effects on organisations. For example, our research reveals that third parties may ostracise those who instigate incivility, which harms cooperation among organisation members and ultimately hinders organisational performance (Howard et al. 2020). Moreover, suffering from workplace ostracism may also push the initial perpetrator of incivility to instigate more uncivilised behaviours in retaliation, leading to a tit-for-tat progression (Ferris et al. 2017; Zhan and Yan 2021). Therefore, organisations should provide explicit provisions to urge employees to comply with workplace etiquette. In addition, relevant complaint channels should be established to encourage employees to resolve disputes in a rational way rather than answering violence with violence.
Consistent with previous studies, our research also shows that witnessing incivility can provoke third parties’ moral anger, prompting them to punish the instigator and provide assistance to the target (Reich and Hershcovis 2015; O’Reilly and Aquino 2011). However, it should be noted that moral anger is more strongly associated with punitive behaviours (e.g., Hill et al. 2025; Jordan and Rand 2020; Skarlicki and Kulik 2005; van Prooijen 2010). If employees are allowed to express their anger by engaging in excessive retaliatory behaviour, it may jeopardise the overall interests of the organisation. This motivates organisations to regularly train employees in emotion regulation to guide them to manage anger in a thoughtful manner. For example, third parties can be encouraged to give priority to helping behaviour (e.g., OCBs) to relieve negative emotions (Reich and Hershcovis 2015).
Finally, we argue that political skill, as an excellent quality of employees, can effectively assist individuals in managing mood swings and rationally resolving unexpected interpersonal conflicts in the workplace. Simultaneously, employees with high political skill maintain friendly relationships with other organisational members through consistently engaging in more OCBs, which is critical to fostering a harmonious organisational climate (Frieder et al. 2019). One plausible recommendation is to incorporate political skill as a vital criterion for employee recruitment and training.
Limitations and future directions
Our research is subject to certain limitations. One primary concern is the potential bias introduced by common method variance due to the reliance on single-source data. Despite this limitation, we assert that the collection of same-source data is essential for achieving our research objectives. We focus on the responses of third parties to specific instigators or targets of incivility, such as their emotional and behavioural responses to the critical incident, which can only be reliably reported by the respondents themselves (Mitchell et al. 2015). In addition, mistreatment in the workplace, such as incivility and ostracism, often manifests in subtle forms, posing challenges for coworkers and supervisors to recognise these deviant behaviours displayed by other colleagues (Ferris et al. 2017; Mitchell et al. 2015). As a result, self-reported sensitive information may yield more precise assessments (Berry et al. 2012). In Study 2, we controlled for common method variance by manipulating the time interval between predictor and criterion variables. However, the 1-week gap we implemented may not adequately minimise the effects of common method variance, as the literature suggests that a minimum of 2 weeks is more effective in reducing this issue (Mitchell et al. 2015; Priesemuth and Schminke 2019).
Second, an exclusive focus on hypothetical and recall techniques may compromise the validity of research findings (Hershcovis et al. 2017). In Study 1, the hypothetical vignettes reflected third parties’ behavioural intentions rather than their actual actions, raising doubts about the external validity of the results. To overcome this limitation, Study 2 utilised a critical incident design that facilitated the investigation of actual responses to witnessed incivility in a real work environment. However, this approach is also retrospective in nature and still suffers from interference such as recall accuracy and social approval. It is recommended that future research adopt a field experimental design that allows for the manipulation of witnessed incivility and enables observation and measurement of third parties’ authentic reactions, thereby enhancing the ecological validity of the study (Reich et al. 2021; Reich and Hershcovis 2015).
Last, several core variables in this research were assessed using tools from Western culture. Variables like workplace ostracism might be quite different in China in their dimensions and manifestations (Ferris et al. 2017). Chinese organisational members place a high value on preserving harmonious interpersonal relationships. Consequently, those who engage in workplace ostracism typically employ more hidden and discreet tactics. This type of ostracism usually does not involve harsh language or obvious behaviour. Instead, it happens quietly to exclude the individual (Su et al. 2021). Furthermore, some scholars have argued that Political Skill Inventory (PSI) may not be completely applicable in the Chinese workplace because this scale fails to capture certain unique aspects of political skill in Chinese culture, such as face-saving and guanxi-maintaining (Liu et al. 2008; Lvina et al. 2012). Additionally, there is also an ongoing debate regarding the cultural applicability of specific sub-dimensions of political skill, particularly apparent sincerity. This is because displaying genuine emotions is viewed as immature and irrational in Chinese culture (Wu and Yang 2019), which differs from individualistic cultures that prioritise authentic expression (Brouer et al. 2015). These issues might explain why our research did not confirm the moderating effect of a political skill sub-dimension. To sum up, researchers must carefully select tools that correspond to the workplace culture of their study in order to assess the cross-cultural validity of their findings in subsequent research.
Conclusion
Our research objectives are achieved, yielding significant insights into third parties’ intervention in workplace incivility. The theoretical model and empirical results suggested that workplace ostracism may serve as a punitive form of intervention by third parties against instigators of incivility, uncovering potential relationships between different types of workplace mistreatment. We found that moral anger played a pivotal role in triggering constructive responses to witnessed incivility, reliably predicting punitive behaviours across two sub-studies. However, the predictive power of moral anger for target-oriented assistance was observed only in Study 1. These findings confirm that moral anger is predominantly directed towards the instigator.
Furthermore, our research delineates the boundary conditions of political skill on third parties’ moral responses to witnessed incivility. Highly politically skilled individuals experienced stronger moral emotions in response to incivility but were more likely to adopt a strategic approach, neither hastily punishing the instigator nor overly advocating for the victim, thus managing the situation strategically. Based on these results, this research answers the current “punishment or assistance primacy” debate regarding third parties’ coping strategy preferences.
This research underscores the need for organisations to create a culture that deters incivility and encourages respect. To attain this goal, we emphasise that organisations must put a premium on third parties’ potential positive contributions when they act as a timely and important force for intervening in workplace incivility. The findings also show the strategic importance of political skill in guiding third parties to rationally cope with negative emotions and engage in beneficial interventions. Therefore, organisational decision makers should appreciate the importance of political skill in developing a healthier and more productive organisational culture for the benefit of employees and the organisation itself.
Data availability
The datasets generated during this study are available in the Humanities and Social Sciences Communications Dataverse under the persistent https://doi.org/10.7910/DVN/ZKNY8P.
References
Ahearn KK, Ferris GR, Hochwarter WA, Douglas C, Ammeter AP (2004) Leader political skill and team performance. J Manag 30(3):309–327. https://doi.org/10.1016/j.jm.2003.01.004
Andersson LM, Pearson CM (1999) Tit for tat? The spiraling effect of incivility in the workplace. Acad Manag Rev 24(3):452–471. https://doi.org/10.5465/AMR.1999.2202131
Arman G (2020) Bystander intervention in the context of abusive supervision: effects of power distance orientation and gender. J Work Organ Psychol 36(2):111–124. https://doi.org/10.5093/jwop2020a11
Artavia-Mora L, Bedi AS, Rieger M (2017) Intuitive help and punishment in the field. Eur Econ Rev 92:133–145. https://doi.org/10.1016/j.euroecorev.2016.12.007
Aryee S, Budhwar PS, Chen ZX (2002) Trust as a mediator of the relationship between organizational justice and work outcomes: test of a social exchange model. J Organ Behav 23(3):267–285. https://doi.org/10.1002/job.138
Aryee EME, Asumeng M, Agyemang CB (2024) Person-environment fit on organizational citizenship behaviour and employee engagement among administrative staff of public universities in Accra, Ghana. J Organ Psychol 24(3). https://doi.org/10.33423/jop.v24i3.7327
Berry CM, Carpenter NC, Barratt CL (2012) Do other-reports of counterproductive work behavior provide an incremental contribution over self-reports? A meta-analytic comparison. J Appl Psychol 97(3):613–636. https://doi.org/10.1037/a0026739
Bevans R (2023) Akaike information criterion | When & how to use it (Example). Scribbr. https://www.scribbr.com/statistics/akaike-information-criterion/
Bowes-Sperry L, O’Leary-Kelly AM (2005) To act or not to act: the dilemma faced by sexual harassment observers. Acad Manag Rev 30(2):288–306. https://doi.org/10.5465/amr.2005.16387886
Brislin RW (1970) Back-translation for cross-cultural research. J Cross-Cult Psychol 1(3):185–216. https://doi.org/10.1177/135910457000100301
Brouer RL, Badaway RL, Gallagher VC, Haber JA (2015) Political skill dimensionality and impression management choice and effective use. J Bus Psychol 30(2):217–233. https://doi.org/10.1007/s10869-014-9344-y
Chen Y, Wang Z, Peng Y, Geimer J, Sharp O, Jex S (2019) The multidimensionality of workplace incivility: cross-cultural evidence. Int J Stress Manag 26(4):356–366. https://doi.org/10.1037/str0000116
Chui CWS, Dietz J (2014) Observing workplace incivility towards women: the roles of target reactions, actor motives, and actor-target relationships. Sex Roles 71:95–108. https://doi.org/10.1007/s11199-014-0367-7
Collier JE (2020) Applied structural equation modeling using AMOS. Routledge, New York. https://doi.org/10.4324/9781003018414
Cortina LM, Hershcovis MS, Clancy KBH (2022) The embodiment of insult: a theory of biobehavioral response to workplace incivility. J Manag 48(3):738–763. https://doi.org/10.1177/0149206321989798
Cortina LM, Magley VJ, Williams JH, Langhout RD (2001) Incivility at the workplace: incidence and impact. J Occup Health Psychol 6(1):64–80. https://doi.org/10.1037/1076-8998.6.1.64
Cortina LM, Kabat-Farr D, Magley VJ, Nelson K (2017) Researching rudeness: the past, present, and future of the science of incivility. J Occup Health Psychol 22(3):299–313. https://doi.org/10.1037/ocp0000089
Coyne I, Gopaul AM, Campbell M, Pankasz A, Garland R, Cousans F (2019) Bystander responses to bullying at work: the role of mode, type and relationship to target. J Bus Ethics 157(3):813–827. https://doi.org/10.1007/s10551-017-3692-2
Dhaliwal NA, Patil I, Cushman F (2021) Reputational and cooperative benefits of third-party compensation. Organ Behav Hum Decis Process 164:27–51. https://doi.org/10.1016/j.obhdp.2021.01.003
Dhanani LY, LaPalme ML (2019) It’s not personal: a review and theoretical integration of research on vicarious workplace mistreatment. J Manag 45(6):2322–2351. https://doi.org/10.1177/0149206318816162
Dhanani LY, LaPalme ML, Joseph DL (2021) How prevalent is workplace mistreatment? A meta‐analytic investigation. J Organ Behav 42(8):1082–1098. https://doi.org/10.1002/job.2534
Faul F, Erdfelder E, Buchner A, Lang AG (2009) Statistical power analyses using G*Power 3.1: tests for correlation and regression analyses. Behav Res Methods 41(4):1149–1160. https://doi.org/10.3758/BRM.41.4.1149
Ferris DL, Chen M, Lim S (2017) Comparing and contrasting workplace ostracism and incivility. Annu Rev Organ Psychol Organ Behav 4:315–338. https://doi.org/10.1146/annurev-orgpsych-032516-113223
Ferris DL, Brown DJ, Berry JW, Lian H (2008) The development and validation of the workplace ostracism scale. J Appl Psychol 93(6):1348–1366. https://doi.org/10.1037/a0012743
Ferris GR, Treadway DC, Perrewé PL, Brouer RL, Douglas C, Lux S (2007) Political skill in organizations. J Manag 33(3):290–320. https://doi.org/10.1177/0149206307300813
Ferris GR, Treadway DC, Kolodinsky RW, Hochwarter WA, Kacmar CJ, Douglas C, Frink DD (2005) Development and validation of the political skill inventory. J Manag 31(1):126–152. https://doi.org/10.1177/0149206304271386
Ferris GR, Ellen III BP, McAllister CP, Maher LP (2019) Reorganizing organizational politics research: a review of the literature and identification of future research directions. Annu Rev Organ Psychol Organ Behav 6:299–323. https://doi.org/10.1146/annurev-orgpsych-012218-015221
Fiori M, Krings F, Kleinlogel E, Reich T (2016) Whose side are you on? Exploring the role of perspective taking on third-party’s reactions to workplace deviance. Basic Appl Soc Psychol 38(6):318–336. https://doi.org/10.1080/01973533.2016.1215984
Folger R (2001) Fairness as deonance. In: Gilliland SW, Steiner DD, Skarlicki DP (eds) Research in social issues in management. Information Age, Greenwich, CT, pp 3–33
Folger R, Stein C (2017) Deonance: expanding the concept. In: Moliner C, Cropanzano R, Martínez-Tur V (eds) Organisational justice: international perspectives and conceptual advances 19. Routledge, London, pp 15–36. https://doi.org/10.4324/9781315648194
Frieder RE, Ferris GR, Perrewé PL, Wihler A, Brooks CD (2019) Extending the meta-theoretical framework of social/political influence to leadership: political skill effects on situational appraisals, responses, and evaluations by others. Pers Psychol 72(4):543–569. https://doi.org/10.1111/peps.12336
Gabriel AS, Chawla N, Rosen CC, Lee YE, Koopman J, Wong EM (2024) Who speaks up when harassment is in the air? A within-person investigation of ambient harassment and voice behavior at work. J Appl Psychol 109(1):39–60. https://doi.org/10.1037/apl0001131
Gummerum M, van Dillen LF, van Dijk E, López-Pérez B (2016) Costly third-party interventions: the role of incidental anger and attention focus in punishment of the perpetrator and compensation of the victim. J Exp Soc Psychol 65:94–104. https://doi.org/10.1016/j.jesp.2016.04.004
Guo G, Gong Q, Li S, Liang X (2021) Don’t speak ill of others behind their backs: receivers’ ostracism (sender-oriented) reactions to negative workplace gossip. Psychol Res Behav Manag 14:1–16. https://doi.org/10.2147/PRBM.S288961
Han S, Harold CM, Oh IS, Kim JK, Agolli A (2022) A meta-analysis integrating 20 years of workplace incivility research: antecedents, consequences, and boundary conditions. J Organ Behav 43(3):497–523. https://doi.org/10.1002/job.2568
Hayes AF (2018) Introduction to mediation, moderation, and conditional process analysis: a regression-based approach, 2nd edn. Guilford Press, New York
Henderson AA, Foster GC, Matthews RA, Zickar MJ (2020) A psychometric assessment of OCB: clarifying the distinction between OCB and CWB and developing a revised OCB measure. J Bus Psychol 35(6):697–712. https://doi.org/10.1007/s10869-019-09653-8
Hershcovis MS, Bhatnagar N (2017) When fellow customers behave badly: witness reactions to employee mistreatment by customers. J Appl Psychol 102(11):1528–1544. https://doi.org/10.1037/apl0000249
Hershcovis MS, Neville L, Reich TC, Christie AM, Cortina LM, Shan JV (2017) Witnessing wrongdoing: the effects of observer power on incivility intervention in the workplace. Organ Behav Hum Decis Process 142:45–57. https://doi.org/10.1016/j.obhdp.2017.07.006
Hill ET, Colquitt JA, Burgess R, Priesemuth M, McClain JT (2025) Third-party perceptions of mistreatment: a meta-analysis and integrative model of reactions to perpetrators and victims. J Appl Psychol 110(1):107–130. https://doi.org/10.1037/apl0001211
Howard MC, Cogswell JE, Smith MB (2020) The antecedents and outcomes of workplace ostracism: a meta-analysis. J Appl Psychol 105(6):577–596. https://doi.org/10.1037/apl0000453
Iqbal M, Yan Y, Shrestha S, Mubarik S, Riaz MN, Jiang L, Nadeem M, Imran MH (2023) A moderated mediation model between perceived incivility and instigated incivility on workplace: a cross-cultural daily diary study. Int J Stress Manag 30(4):388–405. https://doi.org/10.1037/str0000297
Jensen JM, Raver JL (2021) A policy capturing investigation of bystander decisions to intervene against workplace incivility. J Bus Psychol 36(5):883–901. https://doi.org/10.1007/s10869-020-09712-5
Jeong SS, Gong Y, Henderson A (2023) Sympathy or distress? The moderating role of negative emotion differentiation in helping behavior. Asia Pac J Manag 40(4):1429–1458. https://doi.org/10.1007/s10490-022-09819-8
Jiang J, Lu ZR, Zhang W (2011) The development and validation of the workplace ostracism scale. Chin J Clin Psychol 19(6):720–724. https://doi.org/10.16128/j.cnki.1005-3611.2011.06.003
Jordan JJ, Rand DG (2020) Signaling when no one is watching: a reputation heuristics account of outrage and punishment in one-shot anonymous interactions. J Pers Soc Psychol 118(1):57–88. https://doi.org/10.1037/pspi0000186
Kaur N, Kang LS (2023) Perception of organizational politics, knowledge hiding and organizational citizenship behavior: the moderating effect of political skill. Pers Rev 52(3):649–670. https://doi.org/10.1108/PR-08-2020-0607
Kenny DA, Kaniskan B, McCoach DB (2015) The performance of RMSEA in models with small degrees of freedom. Socio Methods Res 44(3):486–507. https://doi.org/10.1177/0049124114543236
Kishton JM, Widaman KF (1994) Unidimensional versus domain representative parceling of questionnaire items: an empirical example. Educ Psychol Meas 54(3):757–765. https://doi.org/10.1177/0013164494054003022
Lin X, Loi R (2021) Punishing the perpetrator of incivility: the differential roles of moral identity and moral thinking orientation. J Manag 47(4):898–929. https://doi.org/10.1177/0149206319870236
Liu H, Jin S, Zhao K (2008) The construct and measurement of organizational political skills in Chinese culture. Chin J Appl Psychol 14(3):220–225. http://www.en.cnki.com.cn/Article_en/CJFDTOTAL-YXNX200803004.htm
Liu P, Li X, Li A, Wang X, Xiong G (2021) How third parties respond to workplace incivility: the moderating role of the dark triad and task interdependence. Pers Individ Differ 171:110427. https://doi.org/10.1016/j.paid.2020.110427
Liu W, Zhou ZE, Che XX (2019) Effect of workplace incivility on OCB through burnout: the moderating role of affective commitment. J Bus Psychol 34(5):657–669. https://doi.org/10.1007/s10869-018-9591-4
Lvina E, Johns G, Treadway DC, Blickle G, Liu Y, Liu J, Atay S, Zettler I, Solga J, Noethen D, Ferris GR (2012) Measure invariance of the political skill inventory (PSI) across five cultures. Int J Cross Cult Manag 12(2):171–191. https://doi.org/10.1177/1470595812439870
McAllister CP, Ellen BP, Ferris GR(2018) Social influence opportunity recognition, evaluation, and capitalization: increased theoretical specification through political skill’s dimensional dynamics J Manag 44(5):1926–1952. https://doi.org/10.1177/0149206316633747
Mehmood S, Rasool M, Ahmed M, Haddad H, Al-Ramahi NM (2024) Role of workplace bullying and workplace incivility for employee performance: mediated-moderated mechanism. PLoS ONE 19(1):e0291877. https://doi.org/10.1371/journal.pone.0291877
Meier LL, Spector PE (2013) Reciprocal effects of work stressors and counterproductive work behavior: a five-wave longitudinal study. J Appl Psychol 98(3):529–539. https://doi.org/10.1037/a0031732
Miner KN, Diaz I, Wooderson RL, McDonald JN, Smittick AL, Lomeli LC (2018) A workplace incivility roadmap: Identifying theoretical speedbumps and alternative routes for future research. J Occup Health Psychol 23(3):320–337. https://doi.org/10.1037/ocp0000093
Mitchell MS, Vogel RM, Folger R (2015) Third parties’ reactions to the abusive supervision of coworkers. J Appl Psychol 100(4):1040–1055. https://doi.org/10.1037/apl0000002
Moon C, Morais C (2022) Understanding the consequences of workplace incivility: the roles of emotional exhaustion, acceptability and political skill. Int J Confl Manag 33(3):425–447. https://doi.org/10.1108/IJCMA-09-2021-0147
Munyon TP, Summers JK, Thompson KM, Ferris GR (2015) Political skill and work outcomes: a theoretical extension, meta-analytic investigation, and agenda for the future. Pers Psychol 68(1):143–184. https://doi.org/10.1111/peps.12066
Ni D, Yang M, Luo S, Chen W (2024) A third-party perspective for examining employee reactions to witnessing coworker incivility towards customers. J Bus Res 177:114616. https://doi.org/10.1016/j.jbusres.2024.114616
O’Reilly J, Aquino K (2011) A model of third parties’ morally motivated responses to mistreatment in organizations. Acad Manag Rev 36(3):526–543. https://doi.org/10.5465/amr.2009.0311
O’Reilly J, Aquino K, Skarlicki D (2016) The lives of others: third parties’ responses to others’ injustice. J Appl Psychol 101(2):171–189. https://doi.org/10.1037/apl0000040
O’Reilly J (2013) Third parties’ morality-motivated response to mistreatment in organizations. Dissertation, University of British Columbia. https://doi.org/10.14288/1.0073884
Podsakoff PM, MacKenzie SB, Podsakoff NP (2012) Sources of method bias in social science research and recommendations on how to control it. Ann Rev Psychol 63:539–569. https://doi.org/10.1146/annurev-psych-120710-100452
Priesemuth M (2013) Stand up and speak up: employees’ prosocial reactions to observed abusive supervision. Bus Soc 52(4):649–665. https://doi.org/10.1177/0007650313490559
Priesemuth M, Schminke M (2019) Helping thy neighbor? Prosocial reactions to observed abusive supervision in the workplace. J Manag 45(3):1225–1251. https://doi.org/10.1177/0149206317702219
van Prooijen JW (2010) Retributive versus compensatory justice: observers’ preference for punishing in response to criminal offenses. Eur J Soc Psychol 40(1):72–85. https://doi.org/10.1002/ejsp.611
Raver JL, Jensen JM, Lee J, O’Reilly J (2012) Destructive criticism revisited: appraisals, task outcomes, and the moderating role of competitiveness. Appl Psychol 61(2):177–203. https://doi.org/10.1111/j.1464-0597.2011.00462.x
Reich TC, Hershcovis MS (2015) Observing workplace incivility. J Appl Psychol 100(1):203–215. https://doi.org/10.1037/a0036464
Reich TC, Hershcovis MS, Lyubykh Z, Niven K, Parker SK, Stride CB (2021) Observer reactions to workplace mistreatment: it’s a matter of perspective. J Occup Health Psychol 26(5):374–392. https://doi.org/10.1037/ocp0000205
Robinson SL, O’Reilly J, Wang W (2013) Invisible at work: an integrated model of workplace ostracism. J Manag 39(1):203–231. https://doi.org/10.1177/0149206312466141
Schilpzand P, De Pater IE, Erez A (2016) Workplace incivility: a review of the literature and agenda for future research. J Organ Behav 37(S1):S57–S88. https://doi.org/10.1002/job.1976
Schoemann AM, Boulton AJ, Short SD (2017) Determining power and sample size for simple and complex mediation models. Soc Psychol Pers Sci 8(4):379–386. https://doi.org/10.1177/1948550617715068
Semmer NK, Tschan F, Jacobshagen N, Beehr TA, Elfering A, Kälin W, Meier LL (2019) Stress as offense to self: a promising approach comes of age. Occup Health Sci 3(3):205–238. https://doi.org/10.1007/s41542-019-00041-5
Skarlicki DP, Kulik CT (2005) Third-party reactions to employee (mis)treatment: a justice perspective. Res Organ Behav 26:183–229. https://doi.org/10.1016/S0191-3085(04)26005-1
Skarlicki DP, Rupp DE (2010) Dual processing and organizational justice: the role of rational versus experiential processing in third-party reactions to workplace mistreatment. J Appl Psychol 95(5):944–952. https://doi.org/10.1037/a0020468
Skitka LJ (2002) Do the means always justify the ends or do the ends sometimes justify the means? A value protection model of justice reasoning. Pers Soc Psychol Bull 28(5):588–597. https://doi.org/10.1177/0146167202288003
Skitka LJ, Mullen E (2002) Understanding judgments of fairness in a real-world political context: a test of the value protection model of justice reasoning. Pers Soc Psychol Bull 28(10):1419–1429. https://doi.org/10.1177/014616702236873
Su T, Chen C, Chen B, Liu J, Ma W (2021) The four deadly sins of workplace ostracism: a meta-analytic examination in Chinese context. Nankai Bus Rev 24(6):106–116. https://doi.org/10.3969/j.issn.1008-3448.2021.06.011
Toomey EC, Rudolph CW, Hannes Z (2021) Age-conditional effects of political skill and empathy on emotional labor: an experience sampling study. Work Aging Retire 7(1):46–60. https://doi.org/10.1093/workar/waaa004
Wang S, Luan K (2024) How do employees build and maintain relationships with leaders? Development and validation of the workplace upward networking scale. J Vocat Behav 150:103985. https://doi.org/10.1016/j.jvb.2024.103985
Wu L, Yang F (2019) Political skill: consequences and theoretical explanation. Adv Psychol Sci 27(12):2109–2121. https://doi.org/10.3724/SP.J.1042.2019.02109
Zhan S, Yan Y (2021) An investigation of the reciprocal effect between workplace incivility and workplace ostracism. Adv Psychol Sci 29(3):560–570. https://doi.org/10.3724/SP.J.1042.2021.00560
Acknowledgements
This work has been supported by the National Social Science Fund of China (Approval ID: Grant 22FGLA006).
Author information
Authors and Affiliations
Contributions
Conceptualisation of the study and research design: Kaichen Zhao, Yu Yan and Zhiqing E. Zhou; Material preparation: all authors; Formal analysis and investigation: Kaichen Zhao; Writing—original draft preparation: Kaichen Zhao; Writing—review and editing: Kaichen Zhao, Yu Yan, Zhiqing E. Zhou, Shuaiping Xiao and Donglu Shan. All authors contributed to the article and approved the submitted version.
Corresponding author
Ethics declarations
Competing interests
The authors declare no competing interests.
Ethical approval
This study was approved by the Ethics Committee of the School of Philosophy, Wuhan University, China (Approval ID: WHU-PSY-2021013; Date: March 1, 2021). The approval scope covers all procedures involving human participants, including experimental design, data collection, and analysis. All procedures adhered strictly to the ethical standards outlined in the Declaration of Helsinki. Participants were provided detailed information about the study’s purpose and procedures prior to obtaining written informed consent. No personally identifiable data (e.g., names, addresses) were collected, with the anonymity and confidentiality of all participants maintained throughout the research process. Individuals retained the right to withdraw participation at any stage without consequence.
Informed consent
Written informed consent was obtained from all participants before each survey. For Study 1, consent was collected on April 4, 2021, and for Study 2, consent was obtained on April 5 and 13, 2022. The consent form clearly stated the study’s purpose, procedures, and confidentiality measures, and emphasised that participation was voluntary. Participants were informed that their anonymity would be strictly preserved through the exclusion of personally identifiable data collection, with all responses stored securely for academic analysis only. They were informed of their right to withdraw at any time without penalty. As all participants were adult employees and no vulnerable groups were involved, no additional consent procedures were required.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary information
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.
About this article
Cite this article
Zhao, K., Yan, Y., Zhou, Z.E. et al. Punish or help? Third parties’ constructive responses to witnessed workplace incivility: the role of political skill. Humanit Soc Sci Commun 12, 516 (2025). https://doi.org/10.1057/s41599-025-04811-1
Received:
Accepted:
Published:
DOI: https://doi.org/10.1057/s41599-025-04811-1
