The ethical considerations section of a research proposal explains how you will protect participants, handle data responsibly, and ensure your study meets institutional and legal standards. It typically appears after your methodology and runs anywhere from one to three pages, depending on how much risk your study involves. While no single PDF template works for every discipline, the section follows a predictable structure grounded in three foundational principles: respect for persons, beneficence, and justice.
Here’s how to build each part of that section, with example language you can adapt for your own proposal.
Start With the Ethical Framework
Open your ethics section by naming the ethical guidelines your study follows. For most research involving human participants, this means the Belmont Report, which establishes three core principles. Respect for persons means people choose freely whether to participate. Beneficence means you maximize benefits and minimize harm. Justice means the burdens and rewards of research are distributed fairly, not concentrated on one group.
If your research is medical or clinical, you should also reference the Declaration of Helsinki. The 2024 revision strengthened protections for vulnerable populations, added requirements for transparency in clinical trials, and expanded commitments to fairness and equity in how research benefits and risks are distributed globally. A single sentence naming these frameworks signals to reviewers that your proposal is grounded in established standards.
Example language: “This study will be conducted in accordance with the ethical principles outlined in the Belmont Report and the 2024 Declaration of Helsinki. Approval will be sought from [institution name] Institutional Review Board prior to data collection.”
Describe Your Informed Consent Process
Informed consent is the most scrutinized part of any ethics section. A valid consent process has three elements: information, comprehension, and voluntariness. Your proposal needs to address all three explicitly.
For the information component, describe what participants will be told before they agree. Federal guidelines require disclosure of specific items: that the study involves research, its purpose, the expected duration of participation, a description of procedures, any foreseeable risks or discomforts, any expected benefits, alternative procedures if applicable, how confidentiality will be maintained, and who to contact with questions or in the event of a research-related injury. For studies involving more than minimal risk, you also need to explain whether compensation or medical treatment is available if something goes wrong.
For comprehension, explain how you’ll confirm participants actually understand what they’re agreeing to. This might mean using plain-language consent forms, reading the form aloud, or asking participants to summarize key points back to you. For voluntariness, state clearly that participation is voluntary, that refusal carries no penalty, and that participants can withdraw at any time without losing any benefits they’re otherwise entitled to.
Example language: “All participants will receive a written consent form explaining the study’s purpose, procedures, potential risks and benefits, and their right to withdraw at any point without consequence. The researcher will review the form verbally with each participant and answer questions before obtaining a signature. Participants will retain a copy of the signed form.”
Outline Risks, Benefits, and Mitigation
Every study carries some level of risk, even if it’s just the inconvenience of filling out a survey. Your job is to identify every foreseeable risk, explain how likely and severe each one is, and describe what you’ll do to reduce it. Then explain the expected benefits, both to participants directly and to the broader field.
Risks fall into several categories: physical (fatigue, discomfort from procedures), psychological (stress, anxiety from interview topics), social (stigma if participation becomes known), economic (lost time or wages), and legal (if data could implicate participants in illegal activity). You don’t need to address every category, only the ones relevant to your study. Be specific. “Minimal risk” is a term reviewers expect to see when risks are no greater than those encountered in everyday life, but you still need to explain what those minimal risks actually are.
Example language: “This study poses minimal risk to participants. The primary risk is mild discomfort when reflecting on personal experiences during interviews. To mitigate this, participants will be informed they may skip any question or end the interview at any time. A list of local counseling resources will be provided to all participants. The anticipated benefit is contributing to improved understanding of [topic], which may inform future interventions for similar populations.”
Explain Data Handling: Confidentiality vs. Anonymity
Reviewers want to know exactly how you’ll protect participant information, and the distinction between confidentiality and anonymity matters here. Anonymity means no one, including you, can link responses to a specific person. The data contains no direct identifiers like names, addresses, birth dates, or IP addresses. True anonymity is only possible in certain designs, like unsigned paper surveys. Face-to-face interviews and phone calls can never be anonymous.
Confidentiality means you can identify participants but commit to protecting that information. Data is either identifiable (labeled with names), coded (labeled with a code that only the research team can link back to a person), or anonymous (no identifying information at all). Most studies use coded data.
Your proposal should specify which approach you’re using, then describe your safeguards. Physical records go in locked cabinets with restricted access. Electronic data gets password protection, unique user credentials, and security software that’s regularly updated. State who will have access, how long data will be retained, and how it will be destroyed when the study concludes.
If your research involves participants in the European Union or California residents, you also need to account for data privacy laws. Under current California regulations, if you retain personal information for longer than 12 months, individuals have expanded rights to access all personal information collected about them going back to January 2022, not just the most recent 12-month window. European regulations carry similar right-to-erasure requirements. Address these obligations if they apply to your participant pool.
Example language: “All data will be coded using participant ID numbers, with the linking key stored separately in a password-protected file accessible only to the principal investigator. Audio recordings will be transcribed within 14 days and the originals permanently deleted. All electronic data will be stored on an encrypted, password-protected drive and destroyed five years after study completion.”
Address Vulnerable Populations
If your study involves children, pregnant women, prisoners, people with diminished decision-making capacity, or any group that may face elevated risk, your ethics section needs additional detail. NIH policy requires individuals of all ages to be included in federally supported research unless there’s a scientific or ethical reason to exclude them, so you may need to justify exclusion as well as inclusion.
For research with children (defined as individuals under 18, or below the legal age of consent in the jurisdiction where the research occurs), you generally need both parental permission and the child’s assent. An IRB will evaluate whether the children in your study are capable of providing assent. If the research involves greater than minimal risk, it’s permitted only under specific conditions that depend on whether participants stand to benefit directly.
For any vulnerable group, describe the specific protections you’ve built in. These might include simplified consent forms, the presence of a legally authorized representative during consent, additional monitoring during the study, or a plan for handling incidental findings.
Example language: “This study includes adolescents aged 14 to 17. Written parental consent and participant assent will be obtained for all minors. Assent forms will use age-appropriate language reviewed by a developmental specialist. Participants will be interviewed in a private room with a parent or guardian present in the adjacent waiting area.”
Identify Your IRB Review Category
Your proposal should indicate what level of IRB review you expect your study to require. There are three categories. Exempt review applies to research that falls into specific low-risk categories and does not require continuing review or the posting of consent forms. Expedited review is handled by a single experienced IRB member and applies to non-exempt research that presents no more than minimal risk and fits within one of nine predefined categories. Full board review (also called convened review) is required for any non-exempt research that doesn’t qualify for expedited review, typically because it involves more than minimal risk.
You don’t decide which category your study falls into; the IRB does. But stating your expected category shows reviewers you understand the regulatory landscape and have designed your study accordingly.
Example language: “This study is expected to qualify for expedited review, as it presents no more than minimal risk and involves only standard interview and survey procedures with adult participants.”
Disclose AI Tool Usage
If you plan to use generative AI tools at any stage of your research, whether for literature screening, data extraction, coding qualitative responses, or drafting portions of your manuscript, your ethics section should say so. Major journals and funding bodies increasingly require explicit disclosure of AI use, and authorship cannot be attributed to AI because it cannot take responsibility for the accuracy or integrity of scientific work.
The key ethical concerns are bias in training data, inaccuracies in AI-generated output, and the risk of fabricated citations. Your proposal should commit to human oversight at every stage where AI is involved, including verifying that all AI-generated citations correspond to real, verifiable sources.
Example language: “AI-assisted tools will be used to support initial screening of literature during the systematic review phase. All AI-generated outputs will be independently verified by two members of the research team for accuracy, completeness, and potential bias. The use of AI tools will be disclosed in all resulting publications.”
Putting the Section Together
A complete ethical considerations section flows in a logical order: framework, consent, risks and benefits, data protection, special populations (if applicable), IRB category, and any additional disclosures like AI use. Each subsection typically runs one to three paragraphs. Use clear subheadings so reviewers can locate specific information quickly.
The section reads strongest when it’s specific to your study rather than generic. Instead of writing “confidentiality will be maintained,” describe exactly how. Instead of “risks will be minimized,” name the risks and the specific steps you’ll take. Reviewers, whether they’re committee members or journal editors, are looking for evidence that you’ve thought through the ethical dimensions of your particular design, not that you’ve memorized a checklist. The example language throughout this article can serve as a starting framework, but adapt every sentence to reflect the actual procedures, population, and context of your research.