What Counts as Workplace Bullying?

Most people who are being bullied at work spend a long time refusing to use the word.

Partly that is because the word sounds like a playground, and adults feel ridiculous applying it to a quarterly review. But there is a better reason for the hesitation, which is that nobody tells you what the threshold is. You are left to decide, privately and without a reference point, whether what is happening to you is serious enough to have a name.

There is a threshold. Researchers have spent thirty years arguing about exactly where it sits, and the version they have converged on is more specific than most people expect. This article sets out that definition, the prevalence data, and the legal position, which is where a great many people are caught off guard.

What this article does not do is tell you whether your particular workplace is a bad place to work overall. That is a broader judgment and it belongs on is your workplace toxic or just difficult. It also does not cover why this conduct is so difficult to evidence, which has its own article at why workplace bullying is hard to prove. Here the question is narrower and prior to both: what actually counts.

What Is the Research Definition?

The definition used across most of the European research tradition comes from Stale Einarsen, Helge Hoel, Dieter Zapf and Cary Cooper, whose formulation has anchored the field for two decades. Bullying at work means harassing, offending or socially excluding someone, or negatively affecting someone's work tasks.

On its own that is broad, so the definition carries four qualifying criteria, and those are what make it usable:

1. Repetition. The person experiences negative acts repeatedly, conventionally at least two, rather than on one occasion.
2. Frequency. Those acts occur weekly or more often.
3. Duration. The pattern persists for six months or longer.
4. Power imbalance. The target has difficulty defending themselves, either because a power disparity existed at the outset or because one developed as the situation escalated.

The fourth criterion is the one people skip, and it is arguably the most important. Einarsen and colleagues treat power disparity as central to the concept. Two colleagues of equal standing in an unpleasant running argument are in a conflict. When one of them can no longer effectively defend themselves, the same behavior has become something else. The escalation into an inferior position is part of the definition rather than a consequence of it.

Why the definition avoids intent

Notice that intent appears nowhere in those four criteria. This is deliberate and it is one of the more useful features of the research definition.

Intent is close to unknowable from the outside. You cannot see into a colleague's reasoning, and any serious attempt to establish what they meant collapses into speculation. The research definition therefore describes what happens and how often, both of which are observable, and leaves motive out. A manager who genuinely believes they are maintaining standards can meet every criterion. So can someone acting with full deliberation. The definition does not distinguish them, because for the purpose of describing what is happening to the target, the distinction does not change much.

Why the Six Month and Weekly Thresholds Exist

The specific numbers look arbitrary, and in a sense they are. Nothing transforms at twenty six weeks. They exist because researchers needed a line that separated bullying from two neighboring things it is genuinely easy to confuse it with.

The first is ordinary workplace conflict. Disagreement between people who work together is normal, sometimes sharp, and not in itself a pathology. Without a duration criterion, every difficult month in every team would qualify, and the term would describe nothing.

The second is incivility. Lynne Andersson and Christine Pearson defined workplace incivility in the Academy of Management Review in 1999 as low intensity deviant behavior with ambiguous intent to harm, in violation of norms for mutual respect. Rudeness, curtness, being talked over. Their central claim was that incivility tends to spiral, escalating through reciprocal exchanges into something more serious. The thresholds mark roughly where a spiral has gone far enough that it is no longer describable as rudeness.

There is an important caveat that the numbers obscure. The criteria were built for research, where you need consistent measurement across thousands of people. They were not built as a permission structure for individuals. Nobody needs to survive twenty five weeks before their experience becomes legitimate, and a single serious incident can be both damaging and actionable without meeting any of these tests. The threshold answers a research question. It does not adjudicate whether you are entitled to be upset.

What Does Not Count, and Why That Matters

A definition earns its keep by excluding things. Several common experiences sit outside this one, and being clear about them is what keeps the term meaningful.

• Reasonable management action, carried out reasonably. Performance feedback, a restructure that removes your favorite part of the job, being told your work is not good enough. Unwelcome is not the test.
• One off incidents, however unpleasant. A single furious outburst can be serious and can breach other policies. It is not bullying under this definition, which is specifically about pattern.
• Mutual conflict between equals. Where both parties can push back with comparable force, the power imbalance criterion is not satisfied.
• Generalized organizational dysfunction. Chronic understaffing and incoherent priorities are real, and genuinely harmful, but they are not directed at a person.

That last exclusion catches a lot of people out, and it is worth separating from the rest. A job can be made unbearable by structure rather than by anyone's conduct. A role built so that it cannot succeed, or one being quietly reduced until the person in it resigns, may involve nobody who is bullying anyone. Our guides to the glass cliff and quiet firing cover two versions of that, and the remedies there are different, because the mechanism is different.

How Common Is It, and Why Do the Numbers Disagree?

If you look this up you will find figures ranging from around one in ten to around one in three, which reasonably makes people suspect that somebody is inventing them. The range is real, and the explanation is methodological rather than sinister.

Morten Nielsen, Stig Berge Matthiesen and Stale Einarsen addressed this directly in the Journal of Occupational and Organizational Psychology in 2010, pooling 102 prevalence estimates from 86 independent samples covering 130,973 people. Mean prevalence varied between roughly 11 and 18 percent, and much of the variation tracked how the question was asked and how the sample was drawn, not just who was asked.

There are two dominant approaches. The self labeling method gives respondents a definition and asks whether it applies to them. The behavioral experience method lists specific negative acts and asks how often each occurred, then applies the criteria afterward. In the 2010 meta analysis, self labeling with a definition supplied produced the lowest average, 11.3 percent, while behavioral checklists produced 14.8 percent. That gap reflects people reporting the experiences while declining the label. Self labeling with no definition given came out highest, at 18.1 percent, and whether a sample was drawn randomly moved the figures by several points as well.

For an individual reader this is more than a technical footnote. The research shows, at scale, that reluctance to call it bullying is the normal response rather than evidence that it is not happening. If you recognize the acts but resist the word, you are behaving exactly like a large share of the people counted in these studies.

The United States picture

In the US, the most widely cited figures come from the Workplace Bullying Institute. Its 2021 U.S. Workplace Bullying Survey, conducted by Zogby Analytics with a national sample of 1,215 adults and a margin of error of plus or minus 2.8 points, found that 30 percent of adult Americans reported direct experience of abusive conduct at work, with 49 percent either directly affected or having witnessed it. Around 4 percent of respondents identified themselves as having bullied someone.

These numbers run higher than the academic meta analysis, and the reason is instructive rather than contradictory. The WBI survey uses a broader construct of abusive conduct and does not impose the six month and weekly thresholds. Both figures are defensible. They are measuring differently calibrated things, which is precisely the point Nielsen and colleagues were making.

Is Workplace Bullying Illegal?

In the United States, generally no, and this surprises almost everyone who has not had reason to find out.

Bullying on its own does not violate Title VII of the Civil Rights Act or any other federal anti discrimination statute. Those laws prohibit harassment that is based on a protected characteristic: race, ethnicity, national origin, religion, disability, sex, sexual orientation or gender identity, pregnancy, genetic information, and age under separate legislation. The conduct becomes unlawful because of what it is based on, not because of how badly the person was treated. The Supreme Court made the same point in Oncale v. Sundowner Offshore Services (1998), warning against turning Title VII into "a general civility code for the American workplace."

The practical consequence is a gap that many people discover at the worst possible moment. A manager who systematically humiliates one person for two years, for reasons unconnected to any protected characteristic, has in most US jurisdictions broken no law. A manager who makes one remark tied to a protected characteristic may have. The severity ordering that most people intuit is not the ordering the statute uses.

Attempts to close this gap have not yet succeeded. The Healthy Workplace Bill, drafted to prohibit abusive work environments without requiring a protected class link, has been introduced in 32 state legislatures, according to the campaign behind the bill. No state has enacted it. In the 2025 to 2026 legislative session it is active in New York, and California and Utah have passed narrower training mandates rather than the bill itself.

Outside the US the position varies considerably, and is often more favorable. Several European countries and Australian jurisdictions have specific rules on bullying or psychological harassment at work. Any of this is worth checking locally rather than assuming, because the differences are large.

Why the legal position still matters if you are not litigating

Most people reading this will never bring a claim, so it would be easy to treat the legal question as academic. It is not, for two reasons.

The first is that it explains a response you may already have encountered. When HR declines to treat a well documented pattern as actionable, that is often not indifference and not a cover up. It reflects a genuine assessment that no legal duty has been triggered. Understanding this in advance makes the response less disorientating, and it changes what you ask for.

The second is that most organizations have internal policies substantially broader than the law requires, covering dignity at work, respect, or conduct. Those policies are usually the applicable framework rather than any statute, and they typically use plainer language than the research definition. Reading your own organization's wording before raising anything is worth more than reading the case law, because the internal wording is what will actually be applied to you.

One closing point about the word itself. A great many people stall for months on whether their experience qualifies, as though the term were a diagnosis requiring certification. It is not. It is a descriptive label that researchers use to study a pattern. Whether your situation meets the six month criterion changes very little about what is happening to you, and nothing about whether you are entitled to take it seriously. If you recognize the acts, you do not have to win an argument with yourself about the vocabulary before you start writing things down.

And if the setting where you recognize the acts is one you cannot afford to walk out of, Corporate Hostage comes at the problem from the practical side rather than the definitional one, as a survival guide for a job you need to keep for now.

And if you are still circling the question of whether your own situation is bad enough to name, the Workplace Reality Check sidesteps the vocabulary entirely and simply asks how often twelve specific things happen in your week.

What Else Do People Ask About Workplace Bullying?

Does it still count if it has been going on for less than six months?

The six month criterion exists so researchers can measure consistently across large samples, not to decide whose experience is legitimate. A shorter pattern can be serious, damaging and worth raising. If you are documenting or reporting, the duration will matter to how it is assessed, which is a reason to start recording early rather than a reason to wait.

Does my manager have to intend to bully me for it to count?

No. The research definition deliberately leaves intent out, because motive cannot be observed from the outside and any attempt to establish it becomes speculation. What is measured is the behavior, its frequency and the power imbalance. A manager who believes they are upholding standards can meet every criterion.

Why do prevalence figures range from 11 percent to 30 percent?

Because the measurement method changes the answer. Asking people to apply a supplied definition to themselves produces lower numbers than listing specific behaviors and asking how often each occurred, while asking about the label with no definition at all produced the highest average of the three. Nielsen and colleagues documented this across 86 samples covering almost 131,000 people. Surveys using broader constructs, such as the Workplace Bullying Institute's, return higher figures again.

Is workplace bullying against the law in the United States?

Usually not on its own. Title VII and related statutes prohibit harassment based on a protected characteristic such as race, sex, religion or disability. Conduct that is severe but unconnected to any protected characteristic generally falls outside them. The Healthy Workplace Bill has been introduced in 32 state legislatures and enacted in none.

If it is not illegal, is there any point raising it internally?

Often yes, because most employers maintain dignity at work or conduct policies that are considerably broader than the legal minimum. Those internal policies, rather than any statute, are usually what will be applied. Reading your organization's own wording before you raise anything tells you more about your position than the law does.

Sources: Einarsen, S., Hoel, H., Zapf, D., and Cooper, C. L., Bullying and Harassment in the Workplace: Theory, Research and Practice (Routledge), setting out the repetition, frequency, duration and power imbalance criteria. Nielsen, M. B., Matthiesen, S. B., and Einarsen, S. (2010), "The impact of methodological moderators on prevalence rates of workplace bullying: A meta-analysis," Journal of Occupational and Organizational Psychology, 83(4), 955-979, pooling 102 estimates from 86 samples (N = 130,973). Workplace Bullying Institute (2021), WBI U.S. Workplace Bullying Survey, conducted by Zogby Analytics, n = 1,215, margin of error plus or minus 2.8 points (workplacebullying.org). Andersson, L. M., and Pearson, C. M. (1999), "Tit for Tat? The Spiraling Effect of Incivility in the Workplace," Academy of Management Review, 24(3), 452-471.

For self-reflection and educational purposes only. Not a substitute for professional mental health support or qualified employment advice. If you feel unsafe, please contact a local support service.