Abusive Supervision: Tepper, Hostile Leadership, Employee Stress, and Workplace Outcomes
Author: Ukrainian Psychological Hub · Published: September 25, 2026 · Editorial Policy
Abusive supervision is a research construct for sustained nonphysical hostility from a supervisor toward a subordinate. In the formulation introduced by Bennett J. Tepper, it refers to subordinates’ perceptions that supervisors repeatedly display hostile verbal and nonverbal behavior while excluding physical contact. That definition gave organizational researchers a specific construct that could be measured separately from broad labels such as a “bad boss,” toxic leadership, workplace bullying, authoritarian management, or simple incompetence (Tepper, 2000; Tepper, 2007).
The word “perceptions” is central. Abusive supervision research has usually measured the subordinate’s experience of a supervisor’s conduct rather than treating hostility as an objectively observed property of the leader. This does not mean the experience is unreal or clinically subjective in the sense of being imagined. It means the construct, as operationalized, combines reported behavior with a target’s interpretation of what the supervisor does. Later reviews argue that this creates a genuine construct-validity problem: researchers need to distinguish abusive supervisory behavior from evaluations of that behavior more carefully (Martinko et al., 2013; Fischer et al., 2021).
The accumulated evidence is substantial enough to reject the idea that abusive supervision is merely a matter of employees disliking strict managers. Meta-analyses and major reviews associate higher perceived abusive supervision with poorer work attitudes and well-being, more turnover intentions and workplace deviance, lower citizenship behavior, and poorer performance-related outcomes, although effect sizes and causal interpretations vary with study design and measurement (Mackey et al., 2017; Tepper et al., 2017). At the same time, a large share of this literature is cross-sectional and based on follower reports, so correlation should not be silently upgraded into proof that every observed outcome was caused by a supervisor.
What Is Abusive Supervision?
The classic definition is narrower than everyday uses of “abusive leadership.” It requires a supervisor–subordinate relationship, a sustained pattern rather than a generic complaint about leadership quality, hostile verbal or nonverbal conduct, and no physical contact within the construct itself. Examples represented in the original measurement tradition include ridicule, public put-downs, hostile reminders of past mistakes, invasion of privacy, the silent treatment, and blaming a subordinate to protect the supervisor’s own image (Tepper, 2000).
Because the construct is tied to the supervisor role, abusive supervision is not identical to leadership psychology as a whole. Leadership is a broader influence process that can be formal or informal; supervision is a formal role relationship involving oversight of work. A supervisor can possess managerial authority without being an influential leader, and an informal leader can influence a group without supervising anyone. This is one reason the distinction between leadership and management matters when interpreting research about workplace hostility.
Abusive supervision is also not a psychiatric diagnosis, personality type, or validated method for inferring a supervisor’s motives. A hostile pattern does not establish Narcissistic Personality Disorder, psychopathy, or any other clinical condition. Research on narcissistic leadership and Dark Triad leadership asks different questions about traits, leader emergence, behavior, and effectiveness. Behavior should be described as behavior unless an independent clinical assessment exists.
Does the Supervisor Have to Intend Harm?
Tepper’s construct is defined through perceived sustained hostility, not through a requirement that researchers prove the supervisor consciously intended to harm the subordinate. This matters because intentions are difficult to observe, supervisors and subordinates may interpret the same episode differently, and hostile behavior can be rationalized as discipline, performance management, humor, urgency, or “toughening someone up.” The empirical task is to measure the conduct and its context rather than read an unobservable motive into every interaction.
Intent can still matter for explanation and organizational response. Deliberate humiliation and a poorly controlled angry outburst are not psychologically identical events. Yet a claim that the supervisor “meant well” does not erase repeated hostile behavior or its consequences. Strong performance management can be direct, specific, demanding, and consequential without ridicule, contempt, intimidation, or degradation.
Does One Bad Incident Count?
A single hostile episode can be serious, especially if it involves threats, discriminatory harassment, violence, or acute safety concerns. The classic abusive-supervision construct, however, emphasizes sustained display. That makes it a pattern construct rather than a label for every isolated conflict. Daily research also shows that leader behavior can fluctuate: in an experience-sampling study, poorer leader sleep quality predicted daily abusive behavior indirectly through self-regulatory depletion, which then related to lower unit engagement (Barnes et al., 2015). Fluctuation does not make a repeated pattern less important; it shows why both chronic levels and day-to-day episodes can matter.
Abusive Supervision vs Toxic Leadership, Destructive Leadership, Bullying, and Incivility
Workplace psychology has accumulated many overlapping labels for harmful interpersonal conduct. Treating them as synonyms makes the literature easier to read and harder to understand. A 2026 integrative review of workplace mistreatment emphasizes that constructs such as abusive supervision, bullying, incivility, aggression, undermining, ostracism, and harassment share a broad mistreatment domain while differing in source, intensity, repetition, intent, and other defining features (Dhanani & Bogart, 2026).
Abusive Supervision vs Toxic Leadership
Toxic leadership is a broader and less consistently standardized label. Different literatures use it to describe damaging leader traits, behaviors, climates, or long-term effects. Abusive supervision is more specific: it has a canonical definition, a dominant measurement tradition, and a large organizational-behavior evidence base centered on supervisor-to-subordinate hostility. A toxic leader may engage in abusive supervision, but the terms should not be substituted automatically.
Abusive Supervision vs Destructive Leadership
Destructive leadership is an umbrella research domain that includes multiple harmful leadership forms. Meta-analytic work links destructive leadership broadly with poorer attitudes, well-being, and performance and with more counterproductive behavior, but the domain contains constructs with different targets and mechanisms (Schyns & Schilling, 2013; Mackey et al., 2021). Abusive supervision is one specific member of that broader family.
Abusive Supervision vs Workplace Bullying
Workplace bullying typically emphasizes repeated negative acts over time and often includes a target’s difficulty defending against them. The aggressor can be a supervisor, coworker, or other workplace actor depending on the definition. Abusive supervision is restricted by source: the actor is a supervisor in relation to a subordinate. The classic abusive-supervision definition also excludes physical contact. The two constructs therefore overlap strongly without being identical.
Abusive Supervision vs Incivility and Interpersonal Conflict
Incivility is usually defined as low-intensity norm violation with ambiguous intent to harm, and it can come from many workplace sources. Interpersonal conflict is broader still and can involve reciprocal disagreement or tension. Abusive supervision is hierarchically asymmetric and centered on sustained hostility from the supervisor. Hershcovis’s influential critique of construct proliferation showed why these categories should be compared carefully rather than multiplied without evidence that they predict meaningfully different outcomes (Hershcovis, 2011).
Abusive Supervision vs Coercive Power
Coercive power is the capacity to impose costs, punish, threaten, or withhold valued outcomes. A supervisor can possess coercive power without acting abusively, because legitimate disciplinary authority may be used according to clear procedures and respectful communication. Conversely, a supervisor can engage in ridicule, humiliation, contempt, or the silent treatment without using a formal punishment. The construct boundary therefore runs between a power resource and a pattern of interpersonal conduct. The same distinction applies to power and control: control over outcomes is not itself evidence of abuse.
Abusive Supervision vs Incompetence and Passive Leadership
A manager may be disorganized, indecisive, technically weak, poor at communication, or unable to coordinate a team. Those failures can damage performance without constituting abusive supervision. Likewise, laissez-faire leadership is characterized by avoidance or absence of needed leadership rather than by sustained hostile behavior. Passive failure and active hostility can coexist, but neither should be inferred from the other.
How Abusive Supervision Is Measured
The field’s dominant instrument is Tepper’s 15-item Abusive Supervision Scale. Employees report how frequently an immediate supervisor engages in behaviors such as ridicule, public put-downs, hostile reminders of mistakes, privacy invasion, or the silent treatment. The original scale used a five-point frequency format and showed strong internal consistency in its development sample. It has since been translated, shortened, adapted, and reused across many countries and industries (Tepper, 2000).
Shorter versions are common, especially the five-item form associated with Mitchell and Ambrose’s work on workplace deviance. But shortening a scale is not a trivial formatting choice. A systematic review found that studies used many different versions of the measure, which complicates comparisons across samples and can change the content domain represented by the score (Fischer et al., 2021). The safest interpretation is therefore behavioral and construct-specific: a higher score indicates more frequent perceived abusive-supervision behaviors as represented by that instrument.
The scale is not a diagnostic test for “having an abusive boss,” and there is no clinical cutoff that converts a workplace relationship into a psychiatric category. It also does not independently prove motive, objective intent, legal harassment, discrimination, or criminal conduct. Those questions require different evidence and, where relevant, different legal or clinical frameworks.
The Measurement Problem: Behavior or Evaluation?
The deepest measurement criticism is conceptual. If abusive supervision is defined as a subordinate’s perception of hostility, and the questionnaire asks the same subordinate to report hostile behaviors, then the score can partly reflect the supervisor’s conduct, the target’s interpretation, and contextual meaning at the same time. Martinko and colleagues proposed separating abusive supervisory behavior from abusive supervisory perceptions; Fischer and colleagues later argued that the field still too often conflates follower evaluation with leader behavior (Martinko et al., 2013; Fischer et al., 2021).
This does not make subordinate reports useless. Targets are often the people with the most direct exposure to private supervisory behavior. It does mean that stronger studies benefit from multiple sources, repeated observations, time separation, behavioral specificity, objective outcomes where available, and designs capable of distinguishing perception from action.
What Does the Evidence Show About Workplace Outcomes?
The strongest conclusion is not that abusive supervision produces one single outcome. It is that perceived abusive supervision sits inside a broad network of adverse employee and organizational correlates. Meta-analytic estimates show relationships with attitudes, well-being, turnover intentions, deviance, citizenship behavior, and performance, but the magnitude depends on the outcome, measure, sample, and design (Mackey et al., 2017).
Stress, Distress, and Well-Being
Tepper’s original study linked abusive supervision with psychological distress, work–family conflict, lower job satisfaction, and lower life satisfaction among employees who remained in their jobs, while also finding a higher likelihood of quitting among employees who perceived more abuse (Tepper, 2000). Later reviews have repeatedly placed stress, burnout, anxiety, negative affect, and other well-being outcomes within the abusive-supervision nomological network (Tepper et al., 2017; Fischer et al., 2021).
The language of stress should stay precise. Workplace stress is not itself a diagnosis, and an association between abusive supervision and distress does not prove that every exposed employee will develop a mental disorder. The evidence supports elevated risk and worse average outcomes, not deterministic prediction for an individual person.
Job Satisfaction, Commitment, and Turnover
Abusive supervision is consistently associated with less favorable job attitudes. In the original research, perceptions of abuse corresponded with lower affective and normative commitment and with greater intentions or likelihood of leaving under some conditions (Tepper, 2000). Reviews and meta-analyses continue to place turnover intentions among the most recurrent outcomes of destructive and abusive leadership (Mackey et al., 2017; Mackey et al., 2021).
Task Performance
Performance effects are real enough to study but easy to overstate. Harris and colleagues used supervisor, self, and formal appraisal measures and found negative relationships for two of three performance indicators, illustrating that the estimated effect depends on how performance is measured (Harris et al., 2007). Meta-analytic work likewise finds generally adverse performance relationships while warning that common-source designs and measurement differences matter (Mackey et al., 2017).
Organizational Citizenship Behavior and Counterproductive Work Behavior
A particularly mature part of the literature concerns discretionary helpful behavior and counterproductive behavior. Mitchell and Ambrose linked abusive supervision with workplace deviance and showed that negative reciprocity beliefs changed the strength of some relationships (Mitchell & Ambrose, 2007). Tepper and colleagues later used a power/dependence framework and found that intentions to quit shaped the relationship between abusive supervision and supervisor- or organization-directed deviance across two studies (Tepper et al., 2009).
Meta-analytic structural equation modeling provides a more integrated mechanism test. Zhang and colleagues found that organizational justice and work stress both mediated relationships between abusive supervision and organizational citizenship behavior (OCB) and counterproductive work behavior (CWB). Justice explained more of the effect on OCB, while work stress explained more of the effect on CWB in their model (Zhang et al., 2019). These are mediation results built from correlational literatures; they strengthen theoretical integration without turning all underlying studies into experiments.
Creativity, Voice, and Learning
Abusive supervision can also interfere with behaviors that require psychological resources and interpersonal risk. A multilevel study by Liu, Liao, and Loi linked abusive supervision with lower employee creativity and traced a cascading process across organizational levels (Liu et al., 2012). The mechanism is plausible because ridicule and humiliation can make speaking up, experimenting, or exposing unfinished ideas more costly. Yet creativity and voice are separate constructs, and neither should be collapsed into psychological safety itself.
Work Unit Engagement
Daily experience-sampling research adds temporal resolution. Barnes and colleagues followed supervisors and subordinates for ten workdays and found an indirect sequence in which poorer leader sleep quality related to ego depletion, more daily abusive supervisory behavior, and lower work-unit engagement (Barnes et al., 2015). This study is useful because it captures within-person variation rather than treating a supervisor as permanently high or low on abuse. It does not imply that sleep deprivation excuses hostility.
Spillover Beyond Work
Workplace mistreatment can travel beyond the supervisor–subordinate dyad. Hoobler and Brass found evidence consistent with displaced aggression: employees who experienced abusive supervision were more likely to undermine family members, linking workplace hostility to behavior at home (Hoobler & Brass, 2006). This spillover perspective helps explain why the cost of abusive supervision cannot be measured only in immediate task output.
Healthcare and Safety-Sensitive Work
Sector-specific evidence matters because consequences can differ when work is interdependent and safety critical. A systematic review of nursing research identified 21 studies and grouped reported consequences into affective, behavioral/performance, relational, cognitive, health/well-being, and nursing-care outcomes. The authors found substantial evidence of harmful workforce effects while noting that evidence about patient safety required further development (Labrague, 2024). A healthcare finding should not be generalized mechanically to every occupation, but it shows why context can change what an outcome means.
Why Might Abusive Supervision Affect Employees?
No single mechanism explains every outcome. The literature has repeatedly used justice, resource depletion and stress, social exchange, power/dependence, identity, affect, and social learning perspectives. The most useful question is therefore not “Which theory is correct?” but “Which mechanism is supported for which outcome and design?”
Organizational Justice
Abuse can communicate that the employee is treated without dignity and that organizational procedures or authorities cannot be trusted to provide fair treatment. Tepper’s original study found that organizational justice mediated many relationships between abusive supervision and employee outcomes (Tepper, 2000). Meta-analytic mediation later showed that justice remained an important explanatory route, especially for organizational citizenship behavior (Zhang et al., 2019).
Stress and Resource Depletion
Hostile supervision can require emotional regulation, vigilance, impression management, avoidance, and recovery. These demands consume attention and self-regulatory resources that could otherwise support work, learning, or recovery. In Zhang and colleagues’ meta-analytic model, work stress was a major route connecting abusive supervision with CWB (Zhang et al., 2019). Resource language should be treated as a mechanism model rather than a claim that every employee has the same fixed “resource tank.”
Social Exchange and Reciprocity
Employment relationships contain expectations of reciprocity. Hostile treatment can weaken the felt obligation to contribute beyond formal requirements and can increase motives to withdraw or retaliate. This helps explain the recurring links with OCB, commitment, and deviance. It also explains why negative reciprocity beliefs can amplify some behavioral responses (Mitchell & Ambrose, 2007).
Power and Dependence
Supervisors often control schedules, assignments, evaluations, information, development opportunities, and other resources. That asymmetry shapes how subordinates respond because retaliation, exit, confrontation, and silence have different costs depending on dependence. Tepper and colleagues explicitly used power/dependence analysis to show that intention to quit altered deviance responses to abusive supervision (Tepper et al., 2009). The broader psychology of power and leadership therefore provides context, while abusive supervision remains its own behavioral construct.
Identity and Social Worth
Ridicule, public humiliation, and contempt can communicate low relational value. Reviews of abusive supervision often describe threats to self-worth, identity, belonging, and respect as plausible pathways into withdrawal, distress, or retaliatory behavior (Tepper et al., 2017). This is psychologically different from ordinary negative feedback, which can be uncomfortable while still affirming the employee’s standing and dignity.
Social Learning and Trickle-Down Effects
Hostility can become part of an organizational climate when people observe that aggressive behavior is modeled, tolerated, or rewarded. Mawritz and colleagues found support for a three-level trickle-down model in which abusive manager behavior related to abusive supervisor behavior, which in turn related to work-group interpersonal deviance; a hostile climate strengthened part of that process (Mawritz et al., 2012). Harmful supervisory behavior can therefore function as a local interaction pattern and as social information about what conduct is permissible.
What Predicts Abusive Supervision?
Antecedents have been studied at several levels. A meta-analysis of 74 studies with more than 30,000 participants organized predictors into supervisor-related, organization-related, subordinate-related, and demographic categories (Zhang & Bednall, 2016). That structure is useful, but it must not be turned into victim blaming. A subordinate characteristic or behavior that statistically predicts more reported abuse does not transfer responsibility for a supervisor’s conduct to the target.
Supervisor-Level Factors
Research has examined supervisor personality, negative affect, self-control, stressors, leadership experiences, and other individual differences. Daily studies suggest that self-regulatory states can matter, as in the leader-sleep research (Barnes et al., 2015). These findings help explain variability in behavior; they do not excuse it. Organizations still decide which conduct is acceptable, how supervisors are selected and trained, and what happens when harmful patterns recur.
Organizational Climate and Cascades
Context can normalize or inhibit hostility. The trickle-down evidence implies that supervisors learn from higher-level managers and from climate cues about acceptable behavior (Mawritz et al., 2012). Systems that reward results while ignoring how results are obtained can create conditions in which abusive behavior survives even when formal values statements prohibit it.
Follower Characteristics and Reverse Causation
Some models include subordinate characteristics, deviance, performance problems, or interaction patterns as antecedents. These findings require unusually careful interpretation. A difficult interaction can be reciprocal, and reverse causation is possible in observational data; Martinko and colleagues explicitly called for models that separate abusive behavior, perceptions, subordinate characteristics, and reciprocal effects (Martinko et al., 2013). None of that converts humiliation, ridicule, or intimidation into an employee-caused event.
What the Evidence Does Not Prove
The abusive-supervision literature is large, yet size alone does not guarantee causal certainty. Fischer and colleagues’ systematic review identified four major problems: conceptual confusion between evaluations and behaviors, measurement limitations, heavy reliance on cross-sectional surveys and vignette experiments, and a low base rate of high abusive-supervision scores that makes extreme-level claims statistically difficult (Fischer et al., 2021).
Correlation Is Not Causation
When the same person rates both supervisor hostility and their own stress or job satisfaction at one point in time, a correlation can reflect real effects, shared method variance, stable response tendencies, reverse influence, or unmeasured context. Longitudinal, multisource, experience-sampling, field, and experimental designs improve inference in different ways, but no single design solves every problem. Statements such as “abusive supervision causes burnout” should therefore be tied to the specific design behind the claim rather than treated as universally established.
Mediation Is Not Automatically a Causal Mechanism
A statistical mediation model can be consistent with a causal process without proving it. The justice and work-stress meta-analytic mediation results are theoretically informative and summarize a very large literature, but many underlying associations are observational (Zhang et al., 2019). Stronger causal claims require temporal ordering, plausible exclusion of alternatives, and designs that test changes in the proposed mediator before changes in the outcome.
There Is No Single Universal Prevalence Rate
Search results often invite a simple answer to “How common is abusive supervision?” The evidence does not support one universal percentage. Estimates vary with occupation, country, scale version, threshold, sampling method, whether employees are asked about any mistreatment or sustained abuse, and whether the study measures average frequency or categorical exposure. The low-base-rate problem highlighted in the 2021 systematic review is an additional reason to avoid presenting one percentage as the prevalence of “abusive bosses” everywhere (Fischer et al., 2021).
Can an Abusive Supervisor Still Be Effective?
An abusive supervisor can produce short-term compliance, meet a deadline, or preside over a high-performing unit for some period. That observation does not demonstrate that abuse caused the performance. High resources, skilled employees, strong systems, selection effects, or fear-based compliance can coexist with hostile treatment. Effectiveness is outcome-specific and time-specific. A leader may hit one performance metric while increasing turnover, silence, deviance, stress, or lost learning.
This distinction parallels the broader separation between leadership emergence, perceived charisma, and leadership effectiveness. Being feared, obeyed, visible, or powerful does not establish effectiveness across the outcomes an organization actually values. The Why Do Leaders Fail? article owns that broader failure intent; abusive supervision focuses specifically on the hostile supervisor–subordinate pattern and its evidence base.
Can Abuse Be “For Performance”?
Supervisors sometimes defend harsh conduct as necessary for standards, urgency, safety, or accountability. The scientific distinction is behavioral. Performance management can specify expectations, document deficiencies, set deadlines, restrict discretion, impose legitimate consequences, and terminate employment without contempt or humiliation. Abusive supervision concerns hostile interpersonal conduct, not the mere existence of authority or negative evaluation.
This distinction also separates abusive supervision from coercive power. A formal warning may be coercive in the sense that it changes the costs of noncompliance, yet it can be procedurally fair and interpersonally respectful. Conversely, ridicule can be abusive even when the supervisor never threatens a formal sanction.
Abusive Supervision, Personality Traits, and the Dark Triad
Research and popular psychology often try to explain hostile managers through narcissism, Machiavellianism, or psychopathy. Trait research can be relevant to antecedents, but behavior and diagnosis must stay separate. The Dark Triad Leadership literature concerns research constructs and leader behavior, while the Narcissistic Leadership literature distinguishes narcissistic traits from Narcissistic Personality Disorder. Observing abusive supervision does not establish either condition.
The same caution applies in reverse: a person can score higher on a personality trait without engaging in abusive supervision. Organizational systems, accountability, selection, role demands, climate, and learned behavior can shape whether dispositional tendencies become enacted conduct. A precise article therefore treats traits as possible predictors or moderators, not as interchangeable names for the behavior itself.
What Employees Can Do
The evidence base describes population-level patterns; an individual employee still has to decide what to do in a specific workplace. A useful first step is behavioral documentation. Record dates, exact statements or actions, context, witnesses, work consequences, and any written communication. Descriptions such as “my supervisor mocked my presentation in front of the team and called my concerns stupid” are more actionable than global labels such as “toxic.”
Second, separate legitimate performance issues from the way they are handled. If the supervisor identifies a real error, document the performance issue and the hostile behavior independently. This prevents the organization from treating a conduct complaint as a dispute about whether feedback was deserved. Both can be true: an employee can need performance correction and still deserve respectful treatment.
Third, use the channels available in the organization: a higher-level manager, HR, an ombuds function, an ethics or speak-up channel, a union representative, or another formal reporting route. Where retaliation, discrimination, threats, physical safety, or legal rights are involved, specialized legal, union, or regulatory advice may be appropriate. The correct channel depends on jurisdiction and employment setting.
Fourth, protect health and functioning. Persistent workplace hostility can affect sleep, stress, concentration, relationships, and decisions about staying. Social support, medical or mental-health care when symptoms warrant it, and practical planning about job mobility can reduce the degree to which a supervisor controls the employee’s options. This is not a prescription to quit; mobility is one factor that changes the power relationship.
What Organizations Can Do
Organizations should design controls around behavior, reporting, evidence, and incentives rather than around labels. Policies can specify prohibited conduct such as humiliation, threats outside legitimate procedures, retaliatory treatment, public ridicule, degrading comments, and intentional exclusion used as punishment. Behaviorally specific rules are easier to investigate than a rule that merely says “do not be toxic.”
Reporting systems should provide more than one route because the direct supervisor is the alleged source of the behavior. Anti-retaliation safeguards, confidential consultation, independent escalation, and documentation standards matter. Repeated low-level reports can reveal a pattern that no single incident shows on its own.
Investigation should also use multiple evidence sources where possible: target accounts, witness accounts, written messages, turnover patterns, climate surveys, prior complaints, scheduling or assignment records, and supervisor explanations. A 360-degree rating can be informative but is not a substitute for an investigation, and a single personality inventory should never be treated as proof of abusive conduct.
Leadership development has a role when the problem is skill, self-regulation, feedback practice, or modeled behavior. It is weaker when organizations use “coaching” as a substitute for accountability after repeated substantiated hostility. Selection, promotion criteria, incentives, manager workload, higher-level role modeling, and consequences for misconduct all shape the local ecology in which abusive supervision occurs.
The trickle-down evidence is especially important for senior leaders. A manager who humiliates direct reports can teach lower-level supervisors that results legitimize hostility. In the three-level study by Mawritz and colleagues, abusive behavior was associated across hierarchical levels and with work-group interpersonal deviance (Mawritz et al., 2012). Culture is transmitted partly through what leaders observe being rewarded or tolerated.
What Supervisors Can Do
Supervisors can reduce risk by making feedback behaviorally specific, private when possible, proportional to the issue, and oriented toward future action. Criticize the work product or decision rather than the person’s worth. Ask questions before attributing motive. Use formal procedures for sanctions. Avoid sarcasm when the power difference makes “jokes” costly to reject. Never use public humiliation as a management technique.
Self-regulation matters because harmful behavior can vary across days and stress states. Sleep loss, overload, conflict, or frustration may increase the probability of poor behavior for some supervisors, but explanation should lead to prevention rather than excuse. Work design, recovery, coaching, and escalation procedures can reduce the chance that a supervisor exports unmanaged stress downward.
When a supervisor has crossed a line, repair requires more than saying the target was “too sensitive.” A credible repair names the behavior, acknowledges the impact, stops repetition, restores fair procedures where possible, and accepts oversight. Because abusive supervision is sustained by patterns, changed behavior over time is more informative than a single apology.
Frequently Asked Questions
What Is the Simplest Definition of Abusive Supervision?
Abusive supervision is a subordinate’s perception that a supervisor repeatedly displays hostile verbal or nonverbal behavior, excluding physical contact, in the classic Tepper definition (Tepper, 2000). The construct is narrower than “bad leadership” and broader than any one hostile act.
What Are Examples of Abusive Supervision?
Common examples in the research tradition include ridicule, public humiliation, contemptuous comments, hostile reminders of mistakes, invasion of privacy, the silent treatment, and shifting blame to protect the supervisor. Whether a specific act meets a workplace policy or legal definition depends on the policy, jurisdiction, and full context.
Is Yelling Once Abusive Supervision?
One episode can be harmful or unacceptable, but the classic construct emphasizes sustained hostile behavior. Serious one-time threats, violence, discriminatory harassment, or safety violations should be treated according to their own standards rather than minimized because the abusive-supervision scale is pattern-based.
Is Abusive Supervision the Same as a Toxic Boss?
No. “Toxic boss” is popular language, and toxic leadership is a broader research label with multiple definitions. Abusive supervision is a specific supervisor–subordinate construct with a well-established measurement tradition.
Is Abusive Supervision the Same as Coercive Leadership?
No. Coercive power concerns control over punishments, threats, or negative outcomes. Abusive supervision concerns sustained hostile interpersonal behavior. They can overlap, but each can occur without the other.
Does Abusive Supervision Cause Burnout and Stress?
Higher perceived abusive supervision is associated with stress, distress, emotional exhaustion, and other well-being outcomes across many studies and reviews. Causal strength varies by design, and a large portion of the literature is observational. The evidence supports a meaningful risk relationship while leaving room for individual and contextual differences (Tepper et al., 2017; Fischer et al., 2021).
Can Abusive Supervision Reduce Performance?
Yes, negative performance relationships appear in primary studies and meta-analytic work, but effect sizes depend on how performance is measured and on study design. Harris and colleagues found negative relationships for formal and supervisor performance ratings but not every performance measure in the same study (Harris et al., 2007).
Can Employees Retaliate?
Research links abusive supervision with counterproductive work behavior and deviance, and power/dependence conditions help explain when retaliation or organization-directed deviance becomes more likely (Mitchell & Ambrose, 2007; Tepper et al., 2009). That descriptive finding does not make retaliation advisable or harmless; it explains a recurring behavioral pathway in the literature.
Is There a Valid Abusive Supervision Test?
There is a widely used research scale: Tepper’s 15-item Abusive Supervision Scale, plus shorter adaptations. It measures reported frequency of abusive-supervision behaviors. It is not a clinical diagnostic test, and the 2021 systematic review raises important construct and measurement concerns that should be considered when interpreting scores (Fischer et al., 2021).
Why Do Some Employees Stay?
People remain in difficult supervisory relationships for many reasons: economic dependence, limited job mobility, professional identity, immigration or credential constraints, benefits, team attachment, career investment, uncertainty, family responsibilities, or hope that conditions will change. Power/dependence research shows why exit options can change responses to abusive supervision (Tepper et al., 2009). Staying is not evidence that the conduct is acceptable.
Can an Organization Prevent Abusive Supervision?
Risk can be reduced through selection, behaviorally specific standards, leader development, multiple reporting channels, anti-retaliation protections, climate monitoring, fair investigations, consequences for repeated misconduct, and senior leaders who do not model hostility. No intervention guarantees elimination, but the literature on antecedents and trickle-down processes shows that abusive supervision is partly organizationally patterned rather than only a private personality problem (Zhang & Bednall, 2016; Mawritz et al., 2012).
The Bottom Line
Abusive supervision is one of the clearest research constructs in the “dark side” of leadership because it has a specific historical definition, a dominant measurement instrument, and a large evidence base. Its core is sustained perceived nonphysical hostility from a supervisor toward a subordinate. The research consistently connects higher abusive-supervision scores with worse employee attitudes, stress and well-being outcomes, turnover intentions, deviance, reduced citizenship behavior, and performance-related costs.
The strongest modern reading is also the most precise. Abusive supervision is not a diagnosis, not a synonym for every difficult manager, not identical to toxic leadership, bullying, incivility, coercive power, or incompetence, and not proven by one unpleasant interaction. Much of the evidence is correlational, measurement remains contested, and extreme levels are difficult to study because they are relatively uncommon in many samples. The field is therefore strongest when it describes behavior carefully, separates constructs, distinguishes association from causation, and evaluates outcomes over time.
Related Articles
References
Barnes, C. M., Lucianetti, L., Bhave, D. P., & Christian, M. S. (2015). “You wouldn’t like me when I’m sleepy”: Leaders’ sleep, daily abusive supervision, and work unit engagement. Academy of Management Journal, 58(5), 1419–1437. https://doi.org/10.5465/amj.2013.1063
Dhanani, L. Y., & Bogart, S. M. (2026). Mapping the mistreatment landscape: An integrative review and reconciliation of workplace mistreatment constructs. Journal of Organizational Behavior, 47(2), 237–260. https://doi.org/10.1002/job.2876
Fischer, T., Tian, A. W., Lee, A., & Hughes, D. J. (2021). Abusive supervision: A systematic review and fundamental rethink. The Leadership Quarterly, 32(6), 101540. https://doi.org/10.1016/j.leaqua.2021.101540
Harris, K. J., Kacmar, K. M., & Zivnuska, S. (2007). An investigation of abusive supervision as a predictor of performance and the meaning of work as a moderator of the relationship. The Leadership Quarterly, 18(3), 252–263. https://doi.org/10.1016/j.leaqua.2007.03.007
Hershcovis, M. S. (2011). “Incivility, social undermining, bullying…oh my!”: A call to reconcile constructs within workplace aggression research. Journal of Organizational Behavior, 32(3), 499–519. https://doi.org/10.1002/job.689
Hoobler, J. M., & Brass, D. J. (2006). Abusive supervision and family undermining as displaced aggression. Journal of Applied Psychology, 91(5), 1125–1133. https://doi.org/10.1037/0021-9010.91.5.1125
Labrague, L. J. (2024). Abusive supervision and its relationship with nursing workforce and patient safety outcomes: A systematic review. Western Journal of Nursing Research, 46(1), 52–63. https://doi.org/10.1177/01939459231212402
Liu, D., Liao, H., & Loi, R. (2012). The dark side of leadership: A three-level investigation of the cascading effect of abusive supervision on employee creativity. Academy of Management Journal, 55(5), 1187–1212. https://doi.org/10.5465/amj.2010.0400
Mackey, J. D., Frieder, R. E., Brees, J. R., & Martinko, M. J. (2017). Abusive supervision: A meta-analysis and empirical review. Journal of Management, 43(6), 1940–1965. https://doi.org/10.1177/0149206315573997
Mackey, J. D., Ellen, B. P., III, McAllister, C. P., & Alexander, K. C. (2021). The dark side of leadership: A systematic literature review and meta-analysis of destructive leadership research. Journal of Business Research, 132, 705–718. https://doi.org/10.1016/j.jbusres.2020.10.037
Martinko, M. J., Harvey, P., Brees, J. R., & Mackey, J. (2013). A review of abusive supervision research. Journal of Organizational Behavior, 34(S1), S120–S137. https://doi.org/10.1002/job.1888
Mawritz, M. B., Mayer, D. M., Hoobler, J. M., Wayne, S. J., & Marinova, S. V. (2012). A trickle-down model of abusive supervision. Personnel Psychology, 65(2), 325–357. https://doi.org/10.1111/j.1744-6570.2012.01246.x
Mitchell, M. S., & Ambrose, M. L. (2007). Abusive supervision and workplace deviance and the moderating effects of negative reciprocity beliefs. Journal of Applied Psychology, 92(4), 1159–1168. https://doi.org/10.1037/0021-9010.92.4.1159
Schyns, B., & Schilling, J. (2013). How bad are the effects of bad leaders? A meta-analysis of destructive leadership and its outcomes. The Leadership Quarterly, 24(1), 138–158. https://doi.org/10.1016/j.leaqua.2012.09.001
Tepper, B. J. (2000). Consequences of abusive supervision. Academy of Management Journal, 43(2), 178–190. https://doi.org/10.5465/1556375
Tepper, B. J. (2007). Abusive supervision in work organizations: Review, synthesis, and research agenda. Journal of Management, 33(3), 261–289. https://doi.org/10.1177/0149206307300812
Tepper, B. J., Carr, J. C., Breaux, D. M., Geider, S., Hu, C., & Hua, W. (2009). Abusive supervision, intentions to quit, and employees’ workplace deviance: A power/dependence analysis. Organizational Behavior and Human Decision Processes, 109(2), 156–167. https://doi.org/10.1016/j.obhdp.2009.03.004
Tepper, B. J., Simon, L., & Park, H. M. (2017). Abusive supervision. Annual Review of Organizational Psychology and Organizational Behavior, 4, 123–152. https://doi.org/10.1146/annurev-orgpsych-041015-062539
Zhang, Y., & Bednall, T. C. (2016). Antecedents of abusive supervision: A meta-analytic review. Journal of Business Ethics, 139(3), 455–471. https://doi.org/10.1007/s10551-015-2657-6
Zhang, Y., Liu, X., Xu, S., Yang, L.-Q., & Bednall, T. C. (2019). Why abusive supervision impacts employee OCB and CWB: A meta-analytic review of competing mediating mechanisms. Journal of Management, 45(6), 2474–2497. https://doi.org/10.1177/0149206318823935