03 · What You Need to Know
Methodological detail matters when it changes the ethical judgment
Ethics and methodology overlap because the way a study is designed determines what participants are asked to undergo and whether those burdens can produce useful knowledge.
The World Medical Association's 2024 Declaration of Helsinki states that medical research involving human participants must have a scientifically sound and rigorous design likely to produce reliable, valid, and valuable knowledge. It also requires the design and performance of the research to be clearly described and justified in a protocol.
WHO's recommended protocol format likewise calls for substantial methodological information where relevant, including study design, population, inclusion and exclusion criteria, interventions, procedures, measurements, instruments, data management, statistical analysis, and safety considerations.
These requirements should not be generalized mechanically to every discipline or committee. They demonstrate why "ethics committees do not care about methodology" is an unsafe rule of thumb.
Start with the study design and what it is supposed to accomplish
The reviewer needs to understand the basic logic of the study. Is it observational or interventional? Experimental or nonexperimental? Cross-sectional or longitudinal? Qualitative, quantitative, mixed methods, or another design? What question or objective is that design intended to address?
The label alone is not enough. Explain the design sufficiently for reviewers to understand why participants are being recruited and why the proposed procedures are needed.
This is also why the ethics committee needs to understand the research design rather than reviewing participant documents in isolation.
Describe participant-facing procedures in concrete detail
If participants experience it, methodological detail about it is usually important.
A reviewer may need to know what participants are asked to do, where procedures occur, how often they occur, approximately how long they take, what interventions or manipulations are involved, what questions are asked, what specimens or measurements are collected, what recordings are made, and what happens during follow-up.
"Participants will undergo the intervention" is plainly inadequate if the intervention determines their exposure to risk. "Participants will complete an online survey" may also be inadequate if the survey asks about trauma, illegal activity, health conditions, or other sensitive matters.
Watch Out
Do not use methodological shorthand to hide ethically important procedures. A familiar research label may mean something precise to you while leaving the reviewer unable to determine what a participant actually experiences.
Explain methodological choices that affect risk
The more directly a methodological choice affects participant welfare, the more important its explanation becomes.
Consider randomization. In some studies, simply stating that participants are randomized would leave major questions unanswered. Reviewers may need to understand the allocation process, study arms, what participants in each arm receive, whether blinding is used, when codes may be broken, and what stopping or safety procedures apply.
The same principle applies outside clinical research. In deception research, reviewers need to understand what information is withheld or altered and how participants are debriefed. In focus groups, the group format creates confidentiality limitations that would not arise in an individual interview. In longitudinal research, repeated contact and retention procedures may introduce privacy considerations absent from a one-time survey.
The method creates the ethical issue, so the method must be described sufficiently to evaluate it.
Explain sampling when participant selection has ethical consequences
Sampling is not purely a statistical or methodological matter. Who is included, excluded, approached, or disproportionately burdened can raise questions of justice and fairness.
Reviewers may therefore need information about the target population, inclusion and exclusion criteria, recruitment source, expected sample, and rationale for selecting particular groups.
Technical sampling calculations may also matter when they bear on scientific validity or unnecessary exposure. For example, an intervention study that recruits substantially more participants than needed could expose additional people to research burdens without adequate justification. A study too small to answer its question may create the opposite problem: participants assume burdens for research unlikely to generate useful knowledge.
Include instrument detail when the instrument itself creates ethical questions
A questionnaire, interview guide, observation protocol, task, stimulus, or other research instrument may reveal ethical issues that a methods summary does not.
Questions about violence, sexuality, illegal behavior, mental health, workplace misconduct, political activity, or other sensitive matters may affect risk and confidentiality. Experimental stimuli may create distress or deception. Observation protocols may raise privacy concerns.
For that reason, some committees require the instruments themselves rather than only a description. Whether surveys, interview guides, and other instruments require approval depends on the applicable review process.
Describe data methods when they affect privacy or confidentiality
Methodology does not stop when data collection ends. How data are linked, transformed, combined, stored, analyzed, and shared may change the privacy and confidentiality risks of the study.
Reviewers may need to know whether identifiers are collected, whether a linkage key exists, whether datasets are combined, whether external services process the data, who has access, and whether quotations, images, recordings, genomic information, geolocation, or other potentially identifying material will be retained or disseminated.
Technical details that materially affect the possibility of identification or disclosure deserve explanation even if participants never see those procedures.
How much analysis detail belongs in the application?
Enough to show that the proposed data can be used to address the research question and, where relevant, that participant burdens are justified by a scientifically credible plan.
WHO's recommended protocol format for health research, for example, calls for statistical methods, sample-size reasoning, treatment of missing data, and sufficiently detailed analysis plans for qualitative research. That does not mean every ethics application in every field needs a complete statistical analysis plan.
Ask what the committee needs to understand. If a particular analytical decision has no meaningful bearing on scientific validity, participant exposure, identifiability, or another ethical issue, an extended technical explanation may be better placed in the full protocol or another methodological document.
Do not confuse technical complexity with ethical relevance
A detail can be technically sophisticated yet ethically unimportant to the review. Another detail can sound mundane but completely change the ethical assessment.
| Methodological detail |
Why it may matter ethically |
Likely priority in the application |
| Who recruits participants |
May affect voluntariness, privacy, or undue influence |
Usually important |
| Duration and frequency of procedures |
Determines participant burden and exposure |
Usually important |
| Randomization and study arms |
Determines what participants receive and associated risks |
Important when applicable |
| Collection of identifiers |
Affects privacy, confidentiality, and data-security risk |
Usually important when identifiable data are involved |
| Highly technical analytic derivation |
May have little effect on participant protection unless it bears on validity or data risk |
Depends on the study |
| Exact formatting conventions used in a software package |
Often does not alter the ethical evaluation |
Usually low priority unless specifically relevant |
The distinction is not between "methods" and "ethics." It is between methodological information that helps the committee make an ethical determination and information that does not.
Complex or higher-risk studies may legitimately need much more detail
Do not force every study into the same level of methodological compression. Clinical trials, invasive procedures, novel interventions, studies involving deception, complex data linkage, research involving people who may require additional protections, and other ethically complicated designs may need extensive explanation.
WHO's protocol guidance, for instance, expects detailed descriptions of interventions and procedures and, for randomized controlled trials, information about randomization, blinding, stopping rules, and conditions for breaking codes.
In these circumstances, detail is not bureaucratic excess. It may be precisely what allows reviewers to determine whether participants are adequately protected.