Can a Researcher Be Responsible for Misconduct Committed by Someone Else on the Team?

If Someone Else Commits Misconduct, Can You Also Be Responsible?

You are part of a research team, and another member fabricates data, falsifies an analysis, or plagiarizes material. Perhaps you are a co-author. Perhaps you supervise that researcher. Perhaps you never handled the questionable data at all.

Are you responsible simply because your name is attached to the project?

Not automatically. Research integrity is a shared responsibility, but misconduct should not be assigned through guilt by association. At the same time, "someone else did it" is not always a complete defense. Your own participation, knowledge, supervisory responsibilities, authorship obligations, response to warning signs, and actions after discovering the problem can create separate accountability.

Another Person’s Misconduct Does Not Automatically Become Yours

Responsibility Should Follow Conduct, Knowledge, and Role

Being on the same research team does not establish misconduct

Imagine that a research assistant secretly invents 30 observations and inserts them into a dataset. The principal investigator, statistician, and three co-authors had no knowledge of the fabrication and no obvious reason to suspect it before the problem was discovered.

The fact that all six researchers worked on the same project does not by itself establish that all six committed fabrication.

Formal misconduct findings require evidence concerning the conduct of the person whose responsibility is being assessed. Under the U.S. Public Health Service framework, for example, research misconduct involves fabrication, falsification, or plagiarism and requires the specified state of mind and other elements for a finding.

The broader principle is that research integrity is shared across a team, but individual culpability should remain traceable.

Direct participation creates the clearest form of responsibility

If you help commit the misconduct, the fact that another person initiated it does not make your own participation disappear.

Suppose a collaborator tells you to remove valid observations because they undermine the preferred conclusion. If you knowingly make those changes and help represent the altered dataset as the complete analysis, the integrity question concerns your conduct as well as the collaborator’s.

The same principle applies when researchers knowingly assist in creating fictitious data, manipulating images, concealing relevant results, or appropriating another person’s work. "I was following instructions" may explain the circumstances, including important power dynamics, but it does not automatically make knowing participation responsible research conduct.

Knowing about misconduct is different from committing it

Knowledge requires a more careful distinction.

Learning that another researcher has committed misconduct does not retroactively make you a participant in the original act. However, what you do with that knowledge can create new responsibilities.

For example, if you discover credible evidence that data underlying a manuscript were fabricated, you may need to preserve relevant records, raise the concern through appropriate channels, prevent continued use of the questionable data where possible, or participate in correcting the research record.

The Singapore Statement emphasizes that researchers should take responsibility for the trustworthiness of their research and report suspected research misconduct to appropriate authorities. It also states that researchers and institutions should respond to irresponsible research practices through appropriate procedures.

Suspicion is not the same as knowledge

Researchers should not be expected to identify misconduct from every unusual result or minor inconsistency.

A surprising dataset may be genuine. An apparent duplicate image may have an innocent explanation. A discrepancy may arise from a software error rather than falsification. Responsible conduct requires appropriate attention to warning signs, but it does not require assuming misconduct whenever research looks odd.

The response should be proportionate to the evidence. Asking for clarification, checking original records, consulting an appropriate expert, or seeking confidential institutional advice may be more appropriate initially than making a public accusation.

Supervisory responsibility is not automatic liability for another person’s misconduct

Principal investigators and supervisors usually have responsibilities for the research environment, training, procedures, and reasonable oversight. The National Academies identifies neglectful or exploitative research supervision as a detrimental research practice, which underscores that supervision itself has integrity implications.

But a supervisor cannot necessarily prevent every concealed act committed by another researcher. A trainee may deliberately circumvent reasonable controls or falsify records in ways the supervisor could not reasonably detect.

Accordingly, two questions should remain separate: did the trainee commit misconduct, and did the supervisor independently fail to meet relevant supervisory responsibilities?

The detailed integrity responsibilities of supervisors toward student research therefore cannot be reduced to automatic liability for everything a student does.

Ignoring warning signs can create a different problem

The analysis changes when a researcher has information indicating a serious integrity risk.

Suppose a principal investigator is repeatedly told that participant records cannot be matched to recruitment logs. The investigator sees evidence that some records may not correspond to real participants but orders the team to continue using the dataset without investigation because the paper is close to submission.

Those facts raise questions about the investigator’s own conduct. Depending on the applicable framework and evidence, the issue might involve reckless misconduct, supervisory failure, another detrimental research practice, or some combination.

The key point is not that the investigator becomes responsible merely because a subordinate committed misconduct. The question is what the investigator knew and did.

Recklessness can matter even without direct fabrication

Under the U.S. Public Health Service framework, qualifying misconduct can be committed intentionally, knowingly, or recklessly. Current Office of Research Integrity guidance explains recklessness in terms of knowledge of a risk of fabrication, falsification, or plagiarism followed by indifference to that risk.

This means that the relevant question may extend beyond who physically changed a file or entered a fabricated observation. Evidence about another researcher’s knowledge and conduct can also matter.

Still, recklessness should not be confused with ordinary negligence. Evidence merely showing that someone should have exercised greater care does not automatically establish the state of mind required under that framework.

Authorship creates accountability without automatically proving misconduct

Co-authorship complicates responsibility because authorship is both credit and accountability.

The National Academies describes authorship as an important mechanism for assigning responsibility for reported work and recommends clearer disclosure of individual author contributions. It also recognizes the challenges created by large and interdisciplinary collaborations in which individual authors may have limited expertise concerning some components of a study.

Being listed as an author therefore does not automatically prove that you participated in a co-author’s misconduct. But authorship is not decorative either. Applicable journal, disciplinary, and institutional standards may require authors to approve the manuscript, take responsibility for specified contributions, and help ensure that integrity concerns affecting the work are appropriately investigated and resolved.

Your exact responsibilities should be assessed under the authorship standards governing the publication rather than inferred solely from author order.

Reasonable reliance on collaborators is often necessary

Complex research depends on distributed expertise. A qualitative researcher may reasonably rely on a statistician for specialized modeling. A clinical researcher may rely on a laboratory scientist for an assay. A physicist may rely on collaborators who operate specialized instrumentation.

Research integrity does not require every collaborator to reproduce every other collaborator’s expertise before accepting their work.

Reasonable reliance has limits, however. If serious inconsistencies appear, explanations repeatedly fail, records cannot be produced, or specific warnings emerge, continued unquestioning reliance may become harder to defend. The appropriate response depends on your role, expertise, access to information, and the seriousness of the warning signs.

Your responsibilities can change after misconduct is discovered

A researcher may be entirely uninvolved in the original misconduct yet become responsible for helping address its consequences.

Suppose you are a co-author who learns after publication that a collaborator fabricated a substantial portion of the dataset. You did not know about the fabrication when the paper was submitted. That fact matters greatly to any assessment of your involvement.

But the publication now contains unreliable evidence. Your responsibility may shift toward cooperating with the investigation, supplying records, communicating accurately with the journal or institution, and helping correct the scholarly record.

A failure to cause the original problem is not a reason to leave a known problem uncorrected.

Responsibility for another person’s act and responsibility for your own omission are different

This distinction helps avoid a common conceptual trap.

The second question can exist even when the answer to the first is no.

One Fabricated Dataset, Different Responsibilities Across the Team

Common Misunderstandings About Responsibility for Someone Else’s Misconduct

Ask What You Did, What You Knew, and What Your Role Required

If misconduct occurs within your team, responsibility should be analyzed more precisely than "we are all responsible" or "it was not my data."

A useful question is: "What responsibility follows from my own role at this point?" That question remains answerable even when you cannot yet determine what another person intended or whether a formal misconduct finding will ultimately be made.

If a Team Member May Have Committed Misconduct, Check

Frequently Asked Questions About Team Member Misconduct

Responsibility Is Not Contagious, but It Can Extend Beyond the Person Who Acted

How to Cite This Guide

This guide is intended to be read, shared, and used in research, teaching, and academic work. If you draw on its ideas, explanations, or other content, please acknowledge the source by citing the guide. Doing so gives appropriate credit and helps your readers locate the original resource.