How to Do Research in High School: Step by Step

Research is asking a question that has no ready answer and working out what the evidence says. This guide walks through the path in order. If you are wondering how it may read on an application, see does research help college admissions first, then come back here to start.

How do you pick a research question?

Pick a question narrow enough that you could answer it with sources or data you can reach this semester. “Why do people use social media?” is too broad. “How do students at my school say they use their phones during study hall?” is something you could actually measure. A workable question names who or what you are studying, and it cannot be settled with a yes or no.

Before you commit, check three things: whether people have already published on it, whether you can get the data or materials, and whether it involves people, animals or hazardous materials, which brings in the rules covered below. If any answer is no, narrow the question or change it. Changing it early costs a day. Changing it after you have collected data costs weeks.

How do you read what has already been published?

Search the published literature before you design anything, so you know what is already answered. Both main tools are free to search. PubMed is a free resource for biomedical and life sciences literature, run by the US National Library of Medicine, and it does not include full-text articles itself, though it often links to them.^[1] Google Scholar covers scholarly articles, theses, books and abstracts, and you then locate the full document through your library or on the web.^[2]

Keep a simple log as you read: the citation, the question the authors asked, what they found, and how it relates to yours. Review articles are a good starting point because they summarize many studies. When a paper is paywalled, your school or public library may be able to get it for you.

How do you choose a method?

Match the method to your question and to what you can reach. The routes below need no lab.

Your question is about	A method that needs no lab
What is already known	A literature review that organizes published studies
Patterns in the population	Analysis of public data, such as federal datasets
What people think or do	A short survey, after the ethics steps below
Whether a finding holds	A replication of a published method with the materials it describes

For public data, Data.gov is the US government’s open data site,^[9] and the Census Bureau lists downloadable tables and a data interface on its data page.^[10] Write down your method before you collect anything, so you cannot adjust it to fit the result later.

Do you need a mentor to start?

No, but one helps. You can begin a literature review or a public-data analysis alone, and a mentor adds feedback, field knowledge and sometimes lab access. Many students ask a teacher first, then a graduate student or faculty member whose published work matches the question.

If a lab is involved, expect rules. Purdue requires a parent or guardian’s written permission for a minor to be in a covered area, and generally at least two adult employees or graduate students present with the minor.^[6] Yale Chemistry welcomes high school students aged 16 or 17 and treats the placement as educational, with no wages.^[7] USC requires a signed parent consent form before work in any of its research labs.^[8] Rules differ by school and department, so ask the lab what applies.

What ethics rules apply to your project?

If you collect information from people, review and consent come first. For science fair projects, Society for Science says research with human participants must be reviewed and approved by an Institutional Review Board before any interaction with participants begins, and that participation may start only after informed consent.^[5] For participants under 18, it requires the student’s assent and written parental permission, and a survey must be attached to the consent form.^[5]

Those are the science fair rules. A university lab or journal may ask for its own approval, so ask your mentor or teacher what applies before you start. Whatever the venue, report your data honestly, cite what you used, and protect participants’ identities.

Where can you share what you find?

Writing it up is the point where a project becomes something others can read. Share it where you can honestly describe it: a class presentation, a science fair, a school or community write-up, or a journal if the work fits one.

Some preprint servers have entry rules. arXiv requires users to be endorsed before submitting a first paper.^[4] If you send work to a journal, check it first. DOAJ says it accepts only open access journals that follow good publishing practice, and removes journals that stop meeting its criteria,^[3] so a listing there is one check among several. Be wary of any outlet that asks for money to be listed as an author. If you are weighing a paid program instead of a project of your own, our checklist on how to tell if a high school research program is legit lists what to verify first. When you describe the result later, say exactly what it is. Our guide on how to describe research on the Common App shows how to state your role and status accurately.

What should you do next?

Write your question in one sentence, find three published papers on it, and decide whether you can answer it alone or need a mentor. If you want an outside view of how your overall profile reads, the free review on ivy-league.io is built for that.

What do we not know?