03 · What You Need to Know
What Ethical Research Requires in Practice
Research Ethics Begins With Justification
Research ethics asks a deceptively simple question: Is it ethically defensible to conduct this research in this way?
That question is broader than checking whether a protocol contains a consent form. It requires examining the purpose of the research, the people or communities who may be affected, foreseeable benefits and harms, how participants are recruited and treated, how information is handled, and whether the study imposes burdens fairly.
There is no universal checklist that resolves every ethical problem. Different ethical frameworks emphasize somewhat different concepts, and requirements vary across disciplines, institutions, jurisdictions, populations, and research designs. The Belmont Report, for example, identifies respect for persons, beneficence, and justice as fundamental principles for research involving human subjects. Other frameworks, including the World Medical Association's Declaration of Helsinki and the CIOMS International Ethical Guidelines, elaborate additional requirements for particular forms of health-related research.
The practical task is therefore to understand the ethical principles relevant to the research and translate them into defensible decisions rather than treating them as abstract ideals.
The Research Question Itself Can Have Ethical Consequences
Ethical evaluation begins before recruitment or data collection. Researchers should be able to explain why the knowledge they seek is worth pursuing and why the proposed research is a reasonable way to obtain it.
This is particularly important when research exposes people to burdens or risks. CIOMS, for example, treats scientific and social value as an ethical requirement in health-related research involving humans. The 2024 Declaration of Helsinki similarly places the generation of knowledge within a framework that protects participants' health, rights, and interests.
This does not mean that every study must transform society or immediately solve a practical problem. Basic research, replication, exploratory work, methodological studies, and research producing uncertain or null findings may still have legitimate value. The relevant question is whether the anticipated knowledge provides a reasonable justification for conducting the study and for any burdens it creates. The relationship between research value and ethical justification deserves careful consideration rather than a ritual sentence about the study being “beneficial.”
Participants Must Be Treated as Persons, Not Merely Sources of Data
For research involving people, ethical treatment ordinarily includes meaningful respect for their autonomy, rights, welfare, privacy, and legitimate interests. The Belmont Report connects respect for persons with recognizing autonomy while providing additional protection when autonomy is diminished.
Informed consent is one important expression of that respect. Belmont describes informed consent in terms of information, comprehension, and voluntariness. A signed form alone therefore cannot establish that consent was genuinely informed or voluntary.
Context matters. A participant may technically agree while facing pressure from a supervisor, teacher, physician, employer, family member, or other person with influence over them. Dense technical language may disclose information without making it understandable. A recruitment incentive may be appropriate in one setting but raise concerns about undue influence in another.
Ethical reasoning asks what the participant actually experiences, not merely what the paperwork records.
Risks Include More Than Physical Injury
Researchers sometimes associate ethical risk primarily with clinical trials or experiments involving bodily interventions. Yet research can cause other forms of harm.
A disclosure of sensitive data might damage someone's privacy, employment, relationships, reputation, or safety. An interview may produce psychological distress. Poor anonymization can expose identities. Research findings can stigmatize a community even when individual participants remain unidentified. Digital research may create privacy concerns because information that is technically accessible online is not necessarily perceived by its creators as unrestricted research material.
This is why the absence of physical intervention does not automatically establish that a study is low-risk or ethically uncomplicated. Whether research can be unethical without physical harm becomes particularly important in social, behavioral, educational, and data-intensive research.
Risks Should Be Minimized and Reasonable in Relation to the Research
Ethical research does not necessarily mean risk-free research. Some worthwhile research cannot eliminate every possibility of discomfort, disclosure, inconvenience, or harm.
The ethical task is first to identify foreseeable risks, then reduce avoidable ones through the study design and safeguards. Remaining risks must be considered in relation to anticipated benefits, if any, and the importance of the knowledge the research may reasonably produce.
The Belmont Report explicitly connects beneficence with maximizing possible benefits and minimizing possible harms. In U.S. human-subject protections, ethics review similarly considers whether risks have been minimized and whether remaining risks are reasonable in relation to anticipated benefits and the importance of the knowledge expected from the study.
Watch Out
Calling a study “minimal risk” does not make it so. Risk should be assessed from the actual procedures, participants, data, setting, and consequences of the research rather than from the researcher's intention not to cause harm.
The Burdens and Benefits of Research Should Be Fairly Distributed
Justice asks researchers to consider who is recruited, who assumes the inconvenience or risk, and who may benefit from the resulting knowledge.
Participant selection should therefore have a defensible relationship to the research question and design. A population should not be recruited simply because it is convenient, accessible, inexpensive, subordinate to the researcher, or less able to refuse participation.
Fairness does not always mean treating everyone identically. Some research legitimately focuses on a particular population because the research question concerns that population. Ethical justification depends on why inclusion or exclusion occurs and whether the resulting distribution of burdens and potential benefits is defensible.
Good Methods and Ethical Research Are Connected
Methodology and ethics are sometimes treated as separate boxes: researchers discuss validity in one section of a protocol and ethics in another. In practice, the two can intersect.
If participants are asked to contribute time, disclose sensitive information, undergo procedures, or accept risk, a design incapable of answering the research question may expose them to burdens without a reasonable prospect of generating the intended knowledge. Methodological weaknesses can therefore become ethical weaknesses when they undermine the justification for those burdens.
That does not mean every imperfect study is automatically unethical. Research always operates under constraints, and methodological limitations differ in severity. The more useful question is whether a weakness materially undermines the study's ability to produce knowledge sufficient to justify what is being asked of participants or others. This is why poor research design can become an ethical problem, rather than remaining purely a technical concern.
Ethics Is Broader Than Ethics Approval
Independent ethics review is an important safeguard in many forms of research, particularly research involving human participants. Review can identify problems that researchers may overlook, assess protections, and require changes before a study proceeds.
But approval is not a transfer of moral responsibility. Researchers continue making decisions after approval: how they recruit participants, respond to unexpected events, protect data, manage protocol changes, communicate with participants, interpret findings, and decide whether emerging circumstances require the research to change or stop.
For the same reason, ethics approval should not be treated as proof that every subsequent action is ethical. Approval concerns a reviewed protocol under specified conditions. Ethical responsibility continues throughout the actual conduct of the research.
Ethics, Law, Institutional Rules, and Research Integrity Overlap but Are Not Identical
Researchers often encounter several kinds of obligations simultaneously. A study may be subject to law, institutional policy, professional standards, funding requirements, disciplinary norms, ethical principles, and rules governing responsible research conduct.
Ethical justification
Asks whether the research and the decisions made within it are morally defensible in light of relevant principles, interests, risks, benefits, rights, and responsibilities.
Compliance
Asks whether applicable legal, regulatory, institutional, contractual, or procedural requirements have been satisfied.
These can support one another, but they should not be collapsed into the same question. A rule may establish a minimum requirement without settling every ethical issue that arises in practice. Conversely, believing that an action is ethically defensible does not authorize a researcher to ignore applicable law or institutional requirements.
Research ethics also overlaps with, but is not synonymous with, research integrity. Questions about responsible authorship, fabrication, falsification, plagiarism, data practices, conflicts of interest, and other dimensions of responsible research conduct may intersect with participant protection while raising additional responsibilities of their own.
Ethical Research Is a Continuing Process
A protocol is written before researchers know exactly how every interaction will unfold. New risks can emerge. Participants may disclose unexpected information. Recruitment may reveal unanticipated power relationships. Data may prove more identifiable than expected. Procedures may change. New evidence can alter the balance between risk and potential benefit.
Ethical research therefore requires continuing judgment. Researchers should recognize when circumstances no longer match the assumptions under which the study was designed or approved and respond through the appropriate ethical, supervisory, or institutional process.
This continuing responsibility is one reason the question “Was my study approved?” is too narrow. The more useful question is “Am I conducting this study in a way that remains ethically defensible?”
