Abstract
This article presents the work of Dr. Manuela Priesemuth. This dissertation examines what happens when employees witness supervisory abuse in the workplace. In particular, it explores whether—and when—employees will respond to witnessing supervisory abuse by engaging in prosocial actions aimed at benefitting the target of abuse. Below, the author discusses the notion of abusive supervision, theoretical perspectives of work on third-party observers, and the conditions under which the author believes third-party observers of abuse are more inclined to engage in positive behavior toward victims. Finally, the reflection commentary provides insights about the research journey in which the author participated.
Abuse, harassment, bullying, or undermining are unfortunately terms we are all too familiar with in organizations as well as society at large. Through research, but also the media, many people are aware of the consequences associated with these types of negative behaviors (Tepper, 2007); that is, we are confronted with the emotional distress victims incur and the measures targets of abuse seek to alleviate distress. In addition to focusing on the strain and negative consequences targets of abuse experience, increasing research suggests incidents of mistreatment rarely happen in isolation (O’Reilly & Aquino, 2011; Skarlicki & Rupp, 2010). Therefore, a focus on observers of abuse and how they may feel and act is warranted.
Specifically, recent events have increased a focus on observer reactions. The public has been outraged about the occurrences involving different coaches and university personnel (e.g., Rutgers, Penn State) who verbally and/or physically abused players or other third parties. Besides being outraged about the unethical actions of the abusers, people also expressed great concern and disappointment about the way the observer and confidants of this unethical behavior reacted. For example, people across the country could not believe that authorities at Penn State did not take the right kind of measures against the observed abusive behavior. These scandals raise questions about the kind of conduct individuals engage in after witnessing abusive behavior of superiors. In particular, people wonder whether and when observers may engage in “morally correct” behavior and take action to benefit the victim of abuse.
This article addresses this important ethical issue in an organizational setting. More specifically, one of the goals is to understand the conditions under which observers of abuse—in particular supervisory abuse—engage in prosocial behavior such as standing up or speaking up (voice) to aid the victim of the abuse. Recently, justice and abuse researchers have expressed an interest in examining third-party reactions to mistreatment at work. In particular, the research provides insights about how third parties—those not directly the target of injustice—react to unfair treatment (Rupp & Bell, 2010; Skarlicki & Kulik, 2005; Skitka, 2002). Research has shown that observing abuse of another person elicits moral unease and deontic reactions in people, such as anger and moral outrage. These deontic reactions prompt people to initiate restorative justice behaviors in which the witness of injustice tends to either punish the transgressor (O’Reilly & Aquino, 2011; Skarlicki & Rupp, 2010; Turillo, Folger, Lavelle, Umphress, & Gee, 2002), refrain from any action at all (Rupp & Bell, 2010), withdraw from the situation, or engage in constructive resistance (Greenbaum, Bardes, Mayer, & Priesemuth, 2009, 2011). This study further extends this line of research and examines the conditions under which an observer of supervisor abuse would exhibit prosocial behavior aimed at benefitting the victim of abuse.
The notion of supervisor abuse in organizations has received substantial attention within the behavioral ethics literature in the past decade (Brown & Mitchell, 2010; Tepper, 2007). Abusive supervision is defined as “subordinates’ perceptions of the extent to which their supervisors engage in the sustained display of hostile verbal and nonverbal behaviors, excluding physical contact” (Tepper, 2000, p. 178). Compelling evidence points to a detrimental effect of abusive supervision on subordinates who are targets of the abuse including decreased levels of job satisfaction, organizational commitment (Tepper, 2000), turnover intentions (Harvey, Stoner, Hochwarter, & Kacmar, 2007), employees’ somatic health issues (Bamberger & Bacharach, 2006), and higher levels of subordinate aggression and deviance (Mitchell & Ambrose, 2007; Thau & Mitchell, 2010).
Beyond increasing our understanding of the conditions under which observers of abuse engage in prosocial behavior to aid the victim of the abuse, this article advances the literature on supervisor abuse in two ways. First, I examine the impact of abusive supervision from a different perspective than is typical in the literature. Rather than focusing on the target of the abuse and the negative impacts of abuse on that specific target, I investigate the effects of abusive supervision on third-party observers. Second, I argue that observing abuse in the workplace may encourage prosocial behavior among employees under distinct individual, relational, and organizational characteristics.
Before delving deeper into the theory development part of the abstract, I will provide some background information about the notion of abusive supervision and its current literature.
Abusive Supervision
As noted at the outset, abusive supervision refers to “subordinates’ perceptions of the extent to which their supervisors engage in the sustained display of hostile verbal and nonverbal behaviors, excluding physical contact” (Tepper, 2000, p. 178); that is, the notion of abusive supervision represents a tyrannical boss who engages in derogatory behavior toward subordinates over a distinct period of time.
In terms of theoretical perspectives, research on abusive supervision has primarily drawn on three distinct literatures to understand the impact an abusive supervisor has on employees. First, social exchange theory (Blau, 1964) and its focus on norms of reciprocity explain employee engagement in deviant behavior in response to abusive supervision. Research found that subordinates reciprocate hostile behavior of their supervisors and retaliate through deviant work behavior (Tepper, Duffy, & Shaw, 2001; Thau, Bennett, Mitchell, & Marrs, 2009).
A second type of theoretical rationale lies in the literature on aggression. Research on displaced aggression provides reasoning why supervisors or employees engage in abusive behavior (Aryee, Chen, Sun, & Debrah, 2007). That is, individuals may redirect aggressive behavior away from the source of harmdoing toward another target (Hoobler & Brass, 2006; Mitchell & Ambrose, 2007). For example, Hoobler and Brass found that employees who perceived their supervisor to be abusive were more prone to engage in aggressive behavior toward their family. This research indicates employees release frustration and aggression toward another target, which is reflected in all kinds of deviant acts, including organizational and interpersonal deviance (Mitchell & Ambrose, 2007).
The third foundation for the abusive supervision literature lies in organizational justice research. In particular, Tepper (2000) proposed a justice-based model of subordinates’ responses to abusive supervision. In order to make fairness assessments in the workplace, individuals draw on perceptions of distributive, procedural, and interactional fairness. Tepper suggests abusive supervision represents a source of all three facets of justice. For example, a supervisor who engages in hostile or abusive behavior toward a subordinate acts against the principles of interactional justice because these actions not only disrespect another person but also provide information suggesting distrust or incapability of that employee. Perceptions of unfairness in turn lead to negative reactions (Greenberg, 1990, 1993).
In all, theories of social exchange, aggression, or fairness provide insights to how targets of abusive supervisory behavior react.
Consequences of Abusive Supervision
Over the past 10 years, much research has focused on the consequences of abusive supervision. In his review of the abusive supervision construct and literature, Tepper (2007) stated that scholars examined consequences of different categories, including work-related attitudes, deviant work behavior, and psychological distress. For example, in the first empirical article about abusive supervision, Tepper (2000) found abusive supervision is negatively related to work attitudes such as job satisfaction and commitment. In addition, he found a positive relationship between abusive supervision and work–family conflict as well as psychological distress. Zellars, Tepper, and Duffy (2002) contributed to research on abusive supervision and organizational citizenship behaviors (OCB), whereas Harris, Kacmar, and Zivnuska (2007) showed a negative impact of abuse on job performance. Finally, work on abuse has also shown a positive relationship with emotional exhaustion (Wu & Hu, 2009) and alcohol abuse (Bamberger & Bacharach, 2006).
Recent work has increased its focus on retaliation and deviant work behaviors (Tepper, Carr, Breaux, Geider, Hu, & Hua, 2009; Thau & Mitchell, 2010). Mitchell and Ambrose (2007) showed that abusive supervision is related to different kinds of workplace deviance, including supervisor-directed deviance, organizational, and nonsupervisor interpersonal deviance.
In sum, research has identified many negative outcomes when working under an abusive supervisor. Therefore, scholars have turned their attention to determine the causes of abusive supervision to find out why supervisors engage in this type of behavior.
Antecedents of Abusive Supervision
One explanation for abusive supervisory behavior may lie in perceptions about the work environment; that is, research has found that supervisors become abusive when they perceive their psychological contract to be breached (Hoobler & Brass, 2006) or when they themselves feel unfairly treated by the organization (Aryee et al., 2007). Drawing on displaced aggression, the supervisor redirects his or her behavior toward other targets, the subordinates. By the same token, Tepper, Duffy, Henle, and Lambert (2006) found that negative justice perceptions of the supervisor lead to depression, which again increased individual tendencies to aggress. Finally, Restubog, Scott, and Zagenczyk (2011) emphasized contextual factors of the work environment as precursors of abusive supervisory behaviors. Specifically, if aggressive norms are present in the workplace, employees tend to model and adapt to these negative norms, which fosters abusive behavior.
Besides examining antecedents and outcomes of abusive supervision, new trends within the literature are emerging. One of these trends lies in examining observer reactions to abuse (Mitchell, Vogel, & Folger, 2012). This dissertation contributes to this emerging stream of research to focus on the question of whether and when witnesses of abuse exhibit prosocial behavior aimed at benefitting the target of abuse.
Observed Abuse and Third-Party Reactions to Injustice
Skarlicki and Kulik (2005) stated that supervisor abuse invokes strong moral emotions in observers, which drive people to act. Thus, recent justice and abuse researchers have expressed an interest in examining third-party reactions to mistreatment. In particular, the justice research provides insights about how third parties react to unfair treatment (Rupp & Bell, 2010; Skarlicki & Kulik, 2005). Most of this research points to retributive reactions of the observer, including punishing the transgressor (Skarlicki & Rupp, 2010).
However, research on the value protection model (VPM), moral mandates (Skitka, 2002; Skitka & Mullen, 2002), and guilt suggests observers of abuse might respond by doing something to aid the target of abuse. Moral mandates are strong attitudes and values that are central to a person’s identity (Skitka, 2002). When these personal moral values are threatened, individuals are morally outraged and inclined to express and protect their values and identity by engaging in cognitive, affective, and/or behavioral adjustments that reinforce their sense of being a good person. Moral cleansing refers to this behavior individuals engage in to reaffirm their sense of being a good person. These behaviors range from simply washing or cleaning one’s hands (Schnall, Benton, & Harvey, 2008; Zhong & Liljenquist, 2006) to prosocial behavior such as human rights activism, working to protect the environment, donating blood, or engaging in silent marches for victims of abuse (Lodewijkx, Kersten, & Van Zomeren, 2008; Skitka & Mullen, 2002).
Research further finds that when people experience survivor guilt, meaning they have gained an accidental advantage over another person (O’Connor, Berry, Weiss, Bush & Sampson, 1997), they attempt to alleviate this guilt by engaging in prosocial behavior toward others (see Baumeister, Stillwell, & Heatherton, 1994, for a review). For example, Wayment (2004) and Wayment, Silver, and Kemeny (1995) found that individuals experiencing survivor guilt after a tragedy engaged in helping behavior and showed support for victims. Finally, Rawlings (1968) found that observers of harm showed the same amount of prosocial behavior as people who felt guilty after engaging in harm doing.
In all, both theoretical and empirical research from the VPM and the guilt literature point to prosocial reactions on the part of individuals who witness abusive supervision.
The Moderating Effects of Individual, Relational, and Organizational Characteristics
The reactions of the witness of abuse might be influenced by individual characteristics of the observer, how the witness perceives the relationship with the target of abuse (relational characteristics), or how the observer perceives their work environment (Figure 1).

Full theoretical model.
Individual Characteristics
When we think about moral courage, specific people come to mind. For example, we think about Martin Luther King, Jr., or individuals who opened their homes to protect victims from the Nazi regime in Europe. In a workplace setting, we consider Cynthia Cooper or Sherron Watkins who decided to “blow the whistle” when their employers (WorldCom and Enron, respectively) were engaging in fraudulent behaviors. All of these people showed moral courage, which is reflected in the acts of individuals who behave morally, even in the face of a threat to themselves. Whistleblowers in organizations, for instance, face tremendous consequences and threats such as humiliation, ridicule, loss of social standing, and unemployment (Kidder & Bracy, 2001), and people engaging in these moral behaviors are thought to be equipped with higher levels of moral courage (Grant, 2002; Jubb, 1999).
My article argues observers of abusive supervision may face threats and consequences if they engaged in behavior supporting the victim. Specifically, supervisors present a higher power in organizations and control important resources for employees. Some abusive supervision and aggression research demonstrated that individuals divert their frustration about supervisors into deviant and aggressive behavior toward others because the risk of future punishment and further retaliation from the supervisor is too daunting (Hoobler & Brass, 2006; Mitchell & Ambrose, 2007; Restubog et al., 2011). To overcome these potentially risky and negative consequences, higher levels of moral courage on the part of the observer may result in greater prosocial behavior after witnessing abuse. In other words, individuals with higher levels of moral courage will be less dissuaded by the potential risk that accompanies aiding the victim of abuse.
Relational Characteristics
In addition to individual characteristics of the observer, relational aspects between the witness of abuse and the victim can also influence the strength of the relationship between observed abusive supervision and prosocial behavior. Specifically, whether the observer perceives close personal ties (friendship) with the victim may provide insights about the strengths of prosocial reactions to witnessing supervisory abuse. Some evidence within the bystander literature and research on altruism point to the fact that mistreatment of people close to oneself elicits prosocial reactions in observers (D’Cruz & Noronha, 2010; Schlenker & Britt, 2001).
Reason for this behavior stems from increased oneness and social identity, prompting observers of abuse to experience a greater sense of shared identity with the victim and to feel more closely related to this individual. This reaction may suggest that observers of abusive supervision experience stronger moral outrage and further strive to alleviate this outrage by engaging in prosocial behavior to benefit the target of abuse. Moreover, observers of abusive supervision may experience stronger feelings of guilt because they have been accidently advantaged over their abused coworker and friend. Research found support for this assertion, such that some of the strongest survivor guilt reactions were found in World War II prison camp survivors, who lost loved ones or close comrades (Neiderland, 1981).
Organizational Characteristics
The importance of justice climate perceptions has been underscored by showing that justice climate represents a significant predictor of employee behavior and attitudes, above and beyond individual-level perceptions of justice (Liao & Rupp, 2005). Hence, the way the observer views overall justice in the workplace may strengthen the relationship between observed abusive supervision and prosocial behavior toward the victim of abuse. Specifically, the organizational context shapes the salience and meaning of organizational events (Kuenzi & Schminke, 2009). In an overall fair work environment, in which employees are treated fairly on a daily basis, individuals may experience stronger moral outrage when witnessing abuse because not only have their moral mandates been violated but also the values of the organization. Furthermore, individuals’ moral mandates, such as treating others with dignity and respect, may also be more salient and instilled in a person who works in an overall fair climate that supports and cherishes these personal values.
The observer of abuse may also experience stronger feelings of guilt. In an overall fair work place, witnesses of abuse may be more likely to perceive an accidental advantage over their coworker. The fact that in a fair work environment, the coworker was chosen as a target of abuse while one was not, may elicit stronger feelings of survivor guilt, which will be alleviated by engaging in prosocial behavior toward the coworker. Overall, stronger moral outrage and violations of moral mandates as well as increased guilt may prompt stronger prosocial reactions toward the victim of abuse.
Findings
In what follows, I will describe the first study from my dissertation and some preliminary results. Specifically, I conducted a laboratory experiment using scenario studies to test some of the hypotheses.
Undergraduate students participated in the data collection effort. The study used a 2 (high observed abusive supervision vs. low observed abusive supervision) X 2 (strong friendship ties vs. weak friendship ties) between-subjects design. Participants were randomly assigned to each condition. The variable of moral courage was assessed as an individual difference prior to reading scenarios describing abuse and friendship ties in the workplace. The scenario instructed students to imagine they were the employee (observer) depicted in the scenario. The scenarios included descriptions of the employee’s immediate supervisor and his abuse toward a coworker. I created the high- and low-observed abusive supervision conditions based on Mitchell and Ambrose’s (2007) distinction between passive-aggressive and active-aggressive abusive behavior. The low-observed abuse condition reflected passive behavior (e.g., not giving credit to the employee for hard and well-performed work), whereas the high-observed abuse condition reflected active abusive behavior (verbally attacking, ridiculing, putting the employee down in front of others). All abusive behaviors depicted in the scenario stem from Tepper’s (2000) abusive supervision scale. The description of the friendship ties in the scenario created a reference to the coworker who was receiving abusive treatment. Either, the participants read about observing mistreatment toward a close friend or simply another coworker.
Measures
Responses for all items were made on a 5-point response scale where 1 = strongly disagree to 5 = strongly agree, unless otherwise noted.
Moral courage
Moral courage was assessed with Sekerka, Bagozzi, and Charnigo’s (2009) six-item measure of endurance of threat, which was taken from the multidimensional scale of professional moral courage. Employees indicated the extent to which they agreed with statements such as “When I encounter an ethical challenge, I take it on with moral action, regardless of how it may pose a negative impact on how others see me,” and “I act morally even if it puts me in an uncomfortable position with my superiors” (α = .73).
Prosocial behavior
In this study, I examine two types of prosocial behaviors in which an observer of abusive supervision could engage. One refers to speaking up or voice behavior and the other refers to standing up behavior. Voice was measured with six items from Van Dyne and LePine’s (1998) scale. The voice items were adapted to reflect the participants’ voice behavior regarding unethical behaviors of the superior and issues affecting the coworker. Sample items include: “I communicate my opinions about unethical situations to others in the work group even if his/her opinion is different and others in the group disagree with me” (α = .71).
I developed six items to reflect the notion of standing up behavior. These items reflect the willingness of an observer to side with or stand up for the target of abuse. Sample items include: “I am willing to stand up for the unfairly treated employee,” and “Regardless of the risk, I will help and support the employee” (α = .94).
Above, I proposed that observed abusive supervision is positively related to prosocial reactions of the witness of abuse. Regression analyses revealed that observed abusive supervision had a direct and positive influence on standing up behavior (β = .21, p < .05) but not on voice behavior.
I suggested that moral courage would moderate the relationship between observed abusive supervision and prosocial reactions of the observer. Regression results revealed no significant interaction effect.
The scenario also examined whether friendship ties moderate the proposed main effect relationship between observed abuse and prosocial reactions. I was able to find partial support for this relationship such that the interaction between observed abuse and friendship ties is significant for standing up behavior (β = .20, p < .01) but not for voice. Plotting the interaction graph (Aiken & West, 1991) shows that, as predicted, under conditions of strong friendship ties the relationship between observed abusive supervision and standing up behavior is strengthened (Figure 2).

Interaction between observed abusive supervision and friendship ties on standing up behavior.
Discussion and Conclusion of Preliminary Findings
Workplace abuse represents a significant ethical challenge for organizations. This dissertation demonstrates that abuse has effects beyond its direct victims and may prompt observers to engage in prosocial behavior aimed at helping the victim. In the study depicted above, it showed these effects are influenced by relational (friendship ties) factors between the witness and target of abuse.
The overall findings of my dissertation go beyond the results of Study 1. For example, two additional field studies I conducted emphasize the importance of other moderating factors. In all, findings from my study extend behavioral ethics research on abusive supervision and third-party injustice in several ways. First and foremost, my findings extend research on abusive supervision by examining observer reactions to abuse instead of focusing on the behaviors the victim of abuse engages in. At the same time, my article extends research on abuse and third-party injustice by showing that individuals do not always engage in retributive (Skarlicki & Rupp, 2010) or withdrawal behavior (Rupp & Bell, 2010; Greenbaum et al., 2011). Instead, results show that under some conditions employees engage in prosocial behavior after witnessing abusive supervision toward a coworker. As shown above, close relational ties between employees matter and are taken into consideration when providing some type of aid to a victim of abuse; that is, when the observer has friendly relationships with the target of abuse, this relationship will impact whether this person is more or less likely to act and display behavior directed to benefit the coworker suffering.
I found these initial results quite hopeful, and I believe my dissertation offers several contributions to understand abusive behavior and how people react to it. I shed light on the conditions under which people are more prone to stand up and help, unfortunately something that is too rare in our current organizations. These findings emphasize the importance of collegiality and the building of personal ties at work. Organizations and employees alike may be able to encourage these types of environments in which stronger bonds between people are formed. In the end, it is these personal ties that could encourage someone to stand up for another person.
Footnotes
Appendix
Declaration of Conflicting Interests
The author declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: The author declares a potential conflict of interest, as some parts of this dissertation abstract may be submitted to other journals for publication purposes.
Funding
The author received no financial support for this research, authorship, and/or publication of this article.
