October 7, 2026

Do High School Researchers Need IRB Approval? A Practical Guide

TL;DR If your project involves other people, you almost certainly need IRB approval, and you need it before you talk to a single participant. Most students who run into trouble here are not being careless. They just did not know the rule existed until after they had collected their data. Working with a mentor who…

TL;DR

If your project involves other people, you almost certainly need IRB approval, and you need it before you talk to a single participant. Most students who run into trouble here are not being careless. They just did not know the rule existed until after they had collected their data. Working with a mentor who has been through ethics review before is the simplest way to avoid that. Horizon Academic Research Program pairs you 1:1 with a PhD scholar or professor for a trimester of guided research, ending in a 20-page paper. Apply here, or come to the next info session first.

You have a research question, you have a survey drafted, and you are about to send it to two hundred students at your school. Stop there for a second. Under science fair rules and under federal research regulations, that survey needs approval before it goes out. This guide walks through exactly how that works for a high school researcher.

Do high school researchers need IRB approval?

Yes, if the project involves human participants. The rule from Society for Science, which governs ISEF and its affiliated fairs, is that student research involving human participants must be reviewed and approved by an Institutional Review Board before any interaction with participants begins. Interaction includes recruitment. Putting a sign-up sheet on a noticeboard counts.

Students often assume IRB review is a university thing that does not apply to them. It applies to you the moment your data comes from people rather than from a database or a lab bench. The board you go through is usually assembled at your school rather than at a university, but the requirement is the same.

In the United States, research with human subjects is governed by federal regulation, specifically 45 CFR 46, commonly called the Common Rule. High school projects are not usually subject to it directly, because it applies to federally funded research and to institutions that have committed to it. Science fair rules exist partly to mirror it, which is why the ISEF process looks so much like a university one.

What counts as human participants research?

The official definition is narrower than most students expect and broader than they hope. A human participant is a living individual about whom you obtain either data or samples through intervention or interaction, or identifiable private information.

Intervention or interaction means you did something to or with the person. You gave them a survey, interviewed them, asked them to taste two drinks, timed them on a task, asked them to use your app. Identifiable private information means you have data about someone that is not public and that can be traced back to them, even if you never met them. Pulling records from a school counsellor’s files would qualify.

A few cases that catch students out:

Your app or website. If anyone other than you tests it, that is human participants research, even if you collect nothing but a satisfaction rating.
Surveys among friends. Small sample, informal setting, still research. There is no minimum number of participants below which the rules switch off.
Online surveys. Posting a form in a Discord server or a subreddit is recruitment. The same approval is needed, and you have the added problem of not knowing whether respondents are minors.
Yourself as a subject. If you are the person being measured in your own study, ISEF treats that as human participants research, and it needs review.
Anything physiological. Exercise trials, sleep tracking, dietary changes, reaction time under caffeine. These need review and often a qualified scientist as well.

Which high school projects are exempt from IRB review?

Four categories are generally exempt under the ISEF human participants rules, and they are worth knowing precisely, because a lot of good projects can be designed to sit inside them.

You test your own invention on yourself only, and it poses no health or safety hazard. The moment you hand it to someone else, exemption ends.
You analyse publicly available data with no interaction with the people in it. Census data, published survey results, open government datasets, the free databases in our guide, all fine.
You observe behaviour in an unrestricted public setting, such as a shopping centre or a park, without interacting with anyone, without manipulating the environment, and without recording anything that identifies an individual.
You receive pre-existing data that is already de-identified or anonymous. Someone else collected it, someone else stripped the identifiers, and you never touch the original.

If your project is currently on the wrong side of that line and the deadline is tight, redesigning into an exempt category is a legitimate move, not a cop-out. A study of published data can be just as strong as a survey study. Our guides to doing psychology research in high school and doing economics research in high school both include designs that avoid participant recruitment entirely.

How do you get IRB approval, step by step?

Here is the full process for a science fair project under ISEF rules. Budget three to four weeks for this, and start before you finalise your instruments, because the board can ask you to change them.

Step 1: Work out whether you need review at all

Read the four exemptions above against your actual design, not your intended design. The test is what you will do with real people, not what your topic is about. A project titled ‘teen anxiety’ that analyses published survey data needs nothing. A project titled ‘plant growth’ that asks fifty classmates to rate leaf colour needs review.

Your adult sponsor is a teacher, mentor, or supervisor who takes responsibility for the safety of your project. They complete Form 1, the Checklist for Adult Sponsor and Safety Assessment Form, which flags what other forms and approvals your project triggers. Pick this person carefully. A sponsor who has supervised fair projects before will catch things you cannot. If you do not have one, our guide to getting research opportunities in high school covers how to approach teachers and outside mentors.

Step 3: Write your research plan on Form 1A

Form 1A, the Student Checklist and Research Plan, is the document your IRB actually reads. It has to describe your question, your methods, your participants, how you will recruit them, what risks they face, and how you will protect their data. Vagueness here is the main reason boards send projects back.

Be specific about four things in particular: who your participants are and how you will reach them, what exactly you will ask or do, what could go wrong for a participant, and where the data will be stored and who can see it. If you have not written a research plan before, our breakdown of the key components of a high school research paper maps closely to what the board wants to see.

Step 4: Assemble your IRB

You need a minimum of three people, and the composition is prescribed rather than up to you. The next section covers exactly who. Nobody on the board may be involved with your project, which rules out your adult sponsor and anyone mentoring the work.

Your board will want to see the actual documents participants will sign, not a description of them. If your participants are under 18, that means a written parental permission form for a legal guardian to sign, plus an assent process for the student themselves. If you are running a survey, the survey itself has to be attached to the consent form, so a parent can read the questions before agreeing.

Write these in plain language. A consent form a fourteen-year-old’s parent cannot understand is not informed consent. Cover what the study is, what participation involves, how long it takes, what the risks are, that participation is voluntary, and that they can stop at any point without consequence.

Step 6: Submit the packet and wait for written approval

Your board reviews the plan and the documents, then signs off on Form 4, the Human Participants form, and Form 1B, the Approval Form. They may approve, approve with changes, or ask for a resubmission. Wait for the signatures. Do not start recruiting while approval is pending, because a project that collected data before the approval date is not fixable later.

Step 7: Collect your data, and keep every piece of paper

Once approved, run the study exactly as described. Keep the signed forms, the consent documents, and the approval itself in one folder, because fairs and journals both ask for them. If anything about your design changes after approval, stop and go back to Step 6 before you resume.

Who has to be on your IRB?

Three members minimum, and each slot has a defined role.

Role	Who this can be	Why they are there
An educator	A teacher at your school who is not your adult sponsor	Judges whether the research plan is coherent and appropriate for a student project
A school administrator	Preferably your principal or vice principal	Takes institutional responsibility for research involving students
A medical or mental health professional	A physician, physician assistant, registered nurse, psychologist, licensed social worker or licensed counsellor	Assesses physical and psychological risk to participants

The medical or mental health professional is the slot students struggle with most. Start there. A school nurse, a counsellor, or a parent of a classmate who happens to be a GP will usually say yes if you ask early and explain that the review is a single meeting. Leaving this to the week before your deadline is how projects miss fairs. Ask in month one.

Which ISEF forms do you need, and when are they due?

The general rule under the ISEF International Rules is that forms are completed and signed before any research takes place. A small set of exceptions are filled in afterwards. Here is the shape of it for a human participants project.

Form	What it is	When
Form 1	Checklist for Adult Sponsor and Safety Assessment	Before research
Form 1A	Student Checklist and Research Plan	Before research
Form 1B	Approval Form	Before research
Form 4	Human Participants	Before research
Informed consent, parental permission and assent documents	Signed by each participant and guardian	Before that person takes part
A copy of your survey or instrument	Attached to the consent form	Before research
Form 3	Risk Assessment, if you are testing an invention or using equipment	Before research
Form 2B	Qualified Scientist, if your project needs specialist oversight	Before research
Form 2C	Regulated Research Institution, if you worked in a university or company lab	After research
Form 7	Continuation Projects, if this builds on last year’s work	After research
Abstract	Summary of the completed project	After research

Download the current versions from the Society for Science forms page rather than reusing a copy from a previous year, because the forms are revised annually and fairs check the version.

Do you need parental permission if your participants are minors?

Yes, and it is two separate things that students often collapse into one. All human participant studies involving minors must receive assent from the student participant and written parental permission from a legal guardian. Two documents, two signatures.

Parental permission is the legal authorisation from a guardian for their child to take part. It has to be in writing, and for a survey it has to include the survey questions.
Assent is the participant’s own agreement, given after the study has been explained to them in language they understand. A minor cannot legally consent, but they can agree, and they can refuse even if their parent has signed.
Either can be withdrawn at any point without the participant having to explain themselves, and your documents should say so plainly.

This matters for practical design. If your participants are classmates, you are collecting permission slips from a hundred sets of parents, which takes time and drops your response rate. Plan for that when you set your sample size, and read up on how sampling methods affect what you can claim before you decide how many responses you actually need.

What happens if you skip IRB approval or change your plan later?

Approval cannot be granted retroactively. There is no form that legitimises data already collected. For a science fair, the practical result is disqualification at the point of screening, and screening happens before judging, so the project never gets seen.

The same applies to changes. If you alter your research plan after approval, the changes have to be reviewed and approved before you resume collecting data. Adding two questions to a survey, widening your age range, switching from anonymous to named responses- all of these need a fresh sign-off. Students usually get caught here because the change felt too small to mention.

Beyond fairs, the consequences are quieter but longer-lasting. Journals that publish high school work ask about ethics approval, and a strong paper with no approvable provenance is one you cannot submit. Journal of Emerging Investigators and Journal of Student Research both expect it for human participants’ work. Our guides to publishing research in high school and research journals for high school students go through what each one asks for.

It also closes off competitions. If you are aiming at science competitions or research competitions later, or planning to turn your paper into a competition or conference submission, the paperwork you file now is what makes that possible.

Does your mentor’s university IRB cover your project?

Sometimes, and it depends on how you are working with them.

You are working inside their approved study. If a professor has an active IRB protocol and you are added to it as a student researcher, you are covered by their approval. You still need Form 2C for a regulated research institution for fair purposes, and that one is completed after the research.
You are running your own study with their guidance. Their protocol does not extend to a project they are not formally running. You need your own review through your school IRB.
You are doing secondary analysis of their existing data. Usually exempt if the data reaches you de-identified, but confirm with them in writing, because ‘de-identified’ has a specific meaning and a spreadsheet with initials and birth dates does not meet it.

This is one of the places where having a mentor who has been through ethics review matters more than their subject expertise. A good mentor tells you in week one that your design will need IRB approval, not in week eight when your survey is written. Our piece on what a research mentor actually does covers the difference, and seven ways to find a research lab is a good place to start if you do not have one yet.

How do you design a project that clears review quickly?

Boards move fastest on projects that are specific, low-risk and clearly bounded. A few habits that help:

Anonymous beats identifiable. If you do not need names, do not collect them. No identifiers means less to protect and a shorter review.
Avoid sensitive topics unless you have proper support. Questions about mental health, trauma, substance use, sexual behaviour or illegal activity raise the risk level sharply and will usually require a qualified scientist. That does not make them off-limits, but it changes the timeline.
Keep the instrument short and finished. Submit the real survey, not a draft with placeholder questions. Boards cannot approve something you are still writing.
Say what you will do with the data when you are done. Deleting raw responses after analysis is a strong answer and costs you nothing.
Pilot with adults first. Testing your survey wording on your parents is not research on human participants, and it catches confusing questions before the board does.

If you are still choosing a direction, how to choose a research topic and broad versus niche research topics both cover feasibility, which is really what ethics review is testing. Subject-specific ideas in psychology, biology, social science, medicine, STEM, AI and history can help you find something workable within your timeline.

Frequently asked questions

Do I need IRB approval for a survey of my classmates?

Yes. Surveys are human participants research regardless of sample size or how well you know the people. You need board approval first, parental permission for anyone under 18, and the survey itself attached to the permission form.

Does my school already have an IRB?

Most schools do not have a standing one. You usually assemble a board for your project, made up of an educator, an administrator and a medical or mental health professional. Larger schools with several fair entries sometimes keep a permanent board, so ask your science department first.

How long does IRB approval take for a high school project?

Plan for three to four weeks from submission to signed approval. The delay is rarely the review itself. It is getting three busy adults into the same room and getting your medical or mental health member to agree to serve.

Can I get IRB approval after I have collected my data?

No. Approval has to come before any interaction with participants, including recruitment, and it cannot be granted retroactively. Data collected before approval cannot be used in an ISEF-affiliated fair.

What if I am the only person in my study?

If you are the subject of your own research, ISEF treats that as human participants research and it needs review. The exception is testing your own invention on yourself where there is no health or safety hazard.

Do online surveys posted on social media need approval?

Yes, and they are harder. You cannot verify whether respondents are minors, which means you cannot guarantee parental permission. Boards often ask you to restrict to adults or to recruit through a channel where ages are known.

Is analysing a published dataset exempt?

Generally yes. Publicly available data with no human interaction is exempt, as is pre-existing data you receive already de-identified. Confirm the dataset is genuinely public and genuinely anonymous before you rely on that.

Do I need this if my project is not going to a science fair?

The ISEF forms are specific to fairs, but the underlying ethical requirement applies to any research involving people, and journals ask about it. Running the process even for a standalone project is worth doing, because it makes the work publishable later.

What is a qualified scientist and when do I need one?

A qualified scientist is an adult with relevant expertise, usually a doctoral degree in the field, who oversees higher-risk work. Form 2B has to be completed by them before any interaction with participants. Studies involving physical or psychological risk usually trigger this.

Does Horizon help with research ethics and IRB?

Yes. Your mentor works through study design with you before you collect anything, which is where ethics questions surface. Apply here or join an info session to see how the trimester is structured.

Resources

External

ISEF rules for human participants research, the primary source for everything in this guide
ISEF International Rules and Guidelines, the full rulebook including risk categories and qualified scientist requirements
Society for Science forms page, download the current year’s Forms 1, 1A, 1B, 3 and 4 here
45 CFR 46, the Common Rule, the federal regulation the fair rules are modelled on
Office for Human Research Protections, plain-language background on informed consent and research with minors
Scribbr research methodology library, for writing up your design clearly on Form 1A
Scribbr on sampling methods, useful when parental permission limits your sample size
Journal of Emerging Investigators and Journal of Student Research, High School Edition, both ask about ethics approval at submission

More from Horizon

Image source: Horizon Academic Research Program

Bridget Carl

Bridget is a Program Director at Horizon Academic. She holds a BS in Public Health and Economics from The George Washington University, where she earned special distinction from her research on the Affordable Care Act individual mandate and open healthcare markets in the US. She also holds a MS in Data Analytics and Policy from Johns Hopkins University where she continued her research focus on public policy and social services, fueled with data science, machine learning and modeling, and cloud computing.

All author posts