High School Research Project Ideas You Can Actually Finish

Choosing a topic is the first step in our guide on how to do research in high school. Most idea lists hand you titles. This page gives you a way to test any idea, then shows where each kind of project gets its data, so you pick one you can finish.

How do you pick a research project idea you can finish?

Test the idea against four questions before you commit: can you state it as one narrow question, can you reach the data or sources, do you have the time, and does it need approval you cannot get. An idea that passes all four is a project. One that fails any of them is a topic to shrink or drop.

Screen	What to check	If it fails
Question	Names who or what you study and what you compare	Cut the population, place or time span
Data	You can open the dataset, archive or papers today	Choose a source you can already download
Time	You can finish a first version in one school term	Drop variables or compare fewer groups
Rules	No people, animals or hazards, or you can get approval	Switch to existing public data

Write the answers down. The “Data” row matters most. If you cannot name where the numbers or documents come from, you do not have a project yet. Our guide to getting a lab position covers the route when you do need a lab, and it comes with its own rules for minors.

Which research project ideas need no lab?

Public datasets and archives give you a project in almost any subject, with no lab and no mentor required to begin. The examples below pair a field with a free source. The questions are hypothetical starting points, not a complete list, and you should narrow each one before you use it.

Field	A starter question (hypothetical)	A free source
Economics	How have prices of one category of goods changed compared with wages in one region?	BLS price, wage and employment databases, plus a data API[5]
Psychology and health	How do reported sleep or screen habits differ across groups of high school students in a national survey?	CDC Youth Risk Behavior Surveillance System datasets, with user guides[6]
Biology without a lab	How do published sequences of one gene compare across related species?	GenBank, which NCBI describes as an annotated collection of publicly available DNA sequences[7]
History and humanities	How did one event appear in the photographs, maps and papers of one period?	Library of Congress collections that are free to use and reuse[8]
Sociology and civics	How do community characteristics differ between two places?	Census Bureau tables, microdata and an API[4]

Data.gov is a catalog of datasets that federal agencies publish, and each agency is responsible for its own data, so read a dataset’s documentation before you rely on it.^[3] Our guide on how to do research in high school explains how to read the published literature around your question.

Do you need ethics approval for your idea?

It depends on whether you collect information from people or animals. If you do, you generally need review before you start. If you only analyze existing public data, the rules are often lighter. 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, and participation may begin only after informed consent.^[1] For participants under 18, it requires the student’s assent and written parental permission.^[1]

The same rules list some projects as exempt from that pre-approval, including reviews of publicly available or published datasets with no human interaction, and public behavioral observation where you do not interact with anyone or record identifying data.^[1] Studies of vertebrate animals must be reviewed and approved before any experiment begins, and observation studies are exempt only if you do not interact with the animals or change their environment.^[2]

These are science fair rules. A school, a university lab or a journal may ask for its own approval, so confirm with your teacher or mentor. If an idea needs approval you cannot get, redesign it around existing data.

How do you narrow a big idea into a small one?

Cut the idea until one sentence names who or what you study, what you compare and where the data comes from. Broad ideas fail because there is too much to read, too many variables and no clear stopping point. Narrow ones can be finished and described honestly.

Work through these steps:

1. Write the broad topic, such as “social media and teenagers.”
2. Choose one group, place or time span.
3. Choose one thing to compare, such as two groups or two periods.
4. Name the dataset or archive you will use, and open it to confirm the variables exist.
5. Rewrite the idea as a question that data could answer.
6. Decide what a first version looks like: a short paper, a chart set or a poster.

“Social media and teenagers” becomes “In one national survey, how do reported habits differ between two groups of respondents?” The narrow version is less impressive to say and far easier to complete. Save the larger question for a later project once you have finished one.

What should you do next?

Pick two ideas from the table above, run each through the four-question screen, and keep the one that passes. Then find three published papers on it before you collect anything. If you want to see how a finished project can read in an application, our guide on how to describe research on the Common App shows how to state your role accurately, and turning a project into a college essay covers the essay. 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?