What Should Researchers Do When They Witness a Serious Problem That Has Nothing to Do With the Research?

What if You See Something Serious You Were Never There to Study?

You are interviewing someone about employment and notice signs that a child in the home may be in danger. During an observational study, you witness serious violence unrelated to your research question. A participant casually mentions that another person is being harmed. During fieldwork, you encounter severe neglect, exploitation, dangerous working conditions, or another situation that was never supposed to become part of the study.

You did not go looking for this information. It may have nothing to do with the research question. Yet once you know about it, can you simply continue taking notes?

There is no universal rule that researchers must intervene whenever they encounter wrongdoing or hardship. But neither does the research role automatically permit researchers to ignore serious threats to people's safety or welfare.

Do Not Assume That “Outside the Study” Means Outside Your Responsibilities

Incidental Problems Can Still Become Research-Ethics Problems

The research question does not define the limits of everything you may encounter

Fieldwork is not a controlled tunnel through which only relevant data enter. Researchers can observe or learn things they never intended to investigate.

UKRI's ESRC guidance explicitly anticipates this possibility. Ethics applications should describe procedures for information arising during fieldwork that causes concern, including disclosures, behaviors, or observed incidents raising significant concerns about the safety or wellbeing of participants or other people affected by the research.

When research involves potentially vulnerable people, ESRC further advises researchers to anticipate situations in which they may acquire responsibilities outside the original project, such as learning that a participant is in significant danger.

The practical implication is important: “That isn't what I am studying” does not by itself resolve the ethical problem.

First distinguish a serious concern from something merely unexpected

Researchers encounter many things they personally dislike, disagree with, or find troubling. Not every questionable decision, family disagreement, poor practice, unhealthy habit, political opinion, or breach of etiquette creates an ethical obligation to intervene.

The threshold rises when there is a credible concern about significant harm. Examples may include immediate threats to life or physical safety, suspected abuse or exploitation, serious neglect, threats toward another person, or circumstances covered by applicable safeguarding or professional reporting obligations.

The relevant categories depend heavily on jurisdiction and professional role. Researchers should avoid creating their own informal legal threshold based on intuition alone.

Immediate danger and non-immediate concerns require different responses

If someone appears to face immediate serious danger, waiting several days for the next research-team meeting may be inappropriate. The project's emergency or safeguarding procedure should identify the appropriate local route for urgent assistance.

A concerning situation that is serious but not immediately dangerous may allow time to consult the principal investigator, safeguarding lead, research governance office, ethics committee, or another designated institutional authority before deciding what action is justified.

Researchers should not delay urgent action merely to protect procedural neatness. Equally, urgency should not be invented where careful consultation is possible.

Confidentiality is strong, but it is not always absolute

Confidentiality is fundamental to many forms of research. Participants may disclose sensitive information precisely because they have been told how their information will be protected.

However, researchers should not promise absolute confidentiality when foreseeable circumstances could require disclosure. ESRC guidance states that when an interview reveals that a participant or another identified person is in significant danger, action may be required to protect that person or third parties. It recommends establishing procedures, systems, and appropriate contacts before such disclosures occur.

If breaching confidentiality is considered necessary, ESRC advises informing the participant where possible unless doing so would increase risk. Any disclosure intended to avert serious harm should be justifiable and directed only to parties empowered to act on the information.

The exact legal position varies. Researchers should therefore distinguish ethical guidance from a legal duty and obtain appropriate institutional or professional advice where necessary.

Consent information should anticipate foreseeable limits to confidentiality

A participant should not first discover the limits of confidentiality after making a serious disclosure if those limits were reasonably foreseeable during study design.

Research involving violence, abuse, safeguarding concerns, serious illegal behavior, or other sensitive matters may need participant information explaining circumstances in which confidentiality cannot be maintained. ESRC guidance specifically asks researchers working with potentially vulnerable participants to consider limits to confidentiality and occasions when those limits may apply.

This does not mean consent materials need an encyclopedic catalogue of every imaginable emergency. They should communicate material limits accurately enough for participants to understand the conditions under which information might be disclosed.

Researchers should not investigate the problem themselves unless that is their role

Seeing something concerning does not automatically authorize the researcher to interrogate people, search for evidence, confront an alleged perpetrator, inspect private records, or conduct an unofficial investigation.

Doing so may increase danger, contaminate evidence, breach privacy, exceed the researcher's competence, or interfere with organizations legally responsible for investigating the concern.

The researcher's task is generally to recognize the issue, protect immediate safety where necessary, document relevant observations appropriately, and activate the designated procedure rather than transforming themselves into investigator, clinician, social worker, security officer, and tribunal before lunch.

Report observations accurately rather than diagnosing what happened

If an incident must be reported, distinguish what you directly observed from what you inferred.

“I saw the adult strike the child twice” conveys different evidence from “the child is being abused.” Likewise, “the participant said they planned to harm X tonight” differs from a general statement that the participant is dangerous.

Factual reporting allows the appropriately authorized person or organization to assess the concern without the researcher overstating what the evidence establishes.

Your professional role may create additional responsibilities

Some researchers are also clinicians, teachers, social workers, psychologists, lawyers, safeguarding professionals, or members of other regulated professions. Their professional obligations may continue to matter while they are conducting research.

The existence and scope of those duties depend on the profession, jurisdiction, setting, and circumstances. Researchers with dual roles should therefore identify possible conflicts between research confidentiality and professional responsibilities before data collection begins.

The phrase “I was acting as a researcher” should not be assumed to suspend duties that continue to apply under professional rules or law.

Intervention can itself cause harm

Acting is not automatically safer or more ethical than not acting. Reporting a concern can expose participants to retaliation, criminalization, family separation, immigration consequences, loss of services, stigma, or other harms. In politically sensitive settings, contacting authorities may itself increase danger.

Researchers should therefore consider who is actually empowered and appropriate to receive information. ESRC guidance cautions that disclosures made to avert serious harm should go only to parties able to act on them.

International research may require particular care because assumptions about police, government agencies, social services, NGOs, or community authorities do not transfer cleanly between settings. ESRC advises international researchers to understand local cultural and political sensitivities and to develop frameworks of responsibility in advance where urgent intervention may arise.

The researcher's own safety still matters

Researchers should not place themselves in serious danger while attempting to intervene. Witnessing violence, abuse, criminal activity, or another emergency can create risks to the observer as well.

If intervention would expose the researcher to immediate danger, the safer course may be to withdraw and contact an appropriate emergency or safeguarding authority from a secure location. The same principle applies when a researcher needs to leave fieldwork because the situation has become unsafe.

The event may require changes to the research

A serious incident can reveal risks that were missing from the original ethics assessment. If one researcher unexpectedly encounters a safeguarding issue, similar encounters may be foreseeable for the rest of the team.

ESRC guidance treats ethics as an ongoing responsibility. Research ethics committees should have procedures for unforeseen events that challenge ethical conduct, and significant concerns can justify changes, further review, suspension, or discontinuation.

The team should therefore ask whether participant information, researcher training, referral arrangements, confidentiality language, escalation procedures, or the research method itself needs revision.

When a Research Visit Reveals Something Outside the Study

Common Mistakes When Researchers Witness Problems Outside the Study

Plan for the Possibility That Research Will Reveal More Than You Asked

If your fieldwork could plausibly expose researchers to serious welfare or safety concerns, decide before data collection how those situations will be handled. This is particularly important in research involving children, vulnerable adults, violence, abuse, exploitation, criminal activity, severe distress, or settings where researchers enter private spaces.

The harder conceptual question is when observation becomes a duty to help. There is no single threshold applicable to every researcher, but serious harm should not be rendered invisible merely because it arrived as an incidental finding.

Are You Prepared to Encounter Serious Problems Outside the Study?

Frequently Asked Questions About Witnessing Serious Problems During Research

Serious Harm Does Not Become Irrelevant Because It Is Incidental

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.