top of page
Post: Blog2_Post

STEM Research Topics for High School Students

Jun 25
16 min read

Updated: Nov 9

Updated November 2026


TL;DR

This post lists 37 STEM research topics for high school students, grouped into biology and health, environmental science, physics and engineering, computer science and data, and chemistry and math. Each topic has a summary of what you would do, who it fits, and what you need to start. Most run on a laptop or household supplies, and the ones that need adult permission or a supervised lab say so. If you want a PhD mentor to help you turn one idea into a college-level paper, you can apply to the Lumiere Research Scholar Program. 


If you're in high school and want to try real research, a STEM project is a good place to start. You could build a model on your laptop and test it against real data, collect water samples from a nearby river, or train a small AI to spot sick plants. This guide gives you 37 ideas in five areas. For each one, you'll see what you would do, who it suits, and what you need to get started. If you want ideas beyond science and tech, see our list of 30 research activities for high school students.


Why should you do STEM research in high school?

Research gets you out of the textbook and into the messy part of science. You pick a question, decide how to test it, collect data, and work out what the results do and don't show. Along the way, you learn to manage a long project and explain your findings in plain language.


A finished project also gives you something specific to talk about in applications and interviews. And it tells you something useful about yourself, because you find out whether you like a field before you commit to it in college. If you're not sure where to begin, our guide to starting out with research in high school covers the first steps.


How do you pick a research topic?

Pick something you can finish. A good first project changes one thing at a time, uses equipment you can get, and fits in a few weeks of regular work. If your idea involves people, ask a teacher or parent for permission before you start, and get consent from every volunteer. If it involves bacteria, chemicals, or heat, do it only with adult supervision.


This table matches common goals to topics in the list below.


If you want to

Look at

Work with data and code, no lab needed

Modeling the Spread of Infectious Diseases; Machine Learning for Predicting Student Performance; AI-Based Image Recognition for Plant Disease Detection; Exploring Neural Networks Through Game Strategy; Sentiment Analysis of Public Text; Predicting Local Temperature With Simple Models; Checking a Public Dataset for Bias; Analyzing Public Transit Delays; How Image Compression Affects Quality; Testing Benford's Law on Real Datasets; Monte Carlo Simulations

Do fieldwork outdoors

Microplastic Pollution in Local Water; Urban Green Spaces and Biodiversity; Backyard Bird Counts and the Weather; Rainwater pH Across Different Storms; Mapping Urban Heat Islands; Soil Erosion on Different Slopes

Build something and test it

Low-Cost Water Filtration System; Renewable Energy Efficiency in Urban vs. Rural Settings; Projectile Motion in Sports; Light Intensity and Solar Cell Efficiency; Temperature and Liquid Viscosity; Air Resistance and Falling Coffee Filters; Truss Design and Bridge Strength; Sound Insulation Materials; Wind Turbine Blade Design

Grow plants or crystals

Soil Composition and Plant Growth; Plant Growth Under Different Colors of Light; Growing Crystals Under Different Conditions

Use a supervised school lab

The Effect of pH on Bacterial Growth; Studying Antibiotic Resistance in Bacteria; The Role of Enzymes in Food Digestion

Study people, with permission and consent

How Sleep Relates to Memory in Teenagers; Spaced Practice vs. Cramming; Heart Rate Recovery After Exercise


Biology, health, and life science

These topics cover disease, microbes, memory, and the human body. If you want more biology ideas, check our list of life sciences research topics.


1. Modeling the Spread of Infectious Diseases

You use the SIR model, which splits a population into Susceptible, Infected, and Recovered groups, to simulate how a disease moves through a community. Set transmission and recovery rates, run the simulation, and compare scenarios with and without vaccination or distancing. You can check your model against real case data.


Best for: You if you like math and want to see equations predict something real.


Requirements: Algebra, a spreadsheet or beginner Python, and a public case-count dataset.


2. The Effect of pH on Bacterial Growth

You grow harmless bacteria in media set to different pH levels and track colony counts or how cloudy the liquid gets over time. Then you compare growth across acidity levels. You can extend it by testing a second species or adding a buffer.


Best for: You if you want hands-on microbiology and careful experimental design.


Requirements: A supervised school or university lab, sterile technique, and non-pathogenic strains approved by your teacher. Don't do this one at home.


3. Studying Antibiotic Resistance in Bacteria

You expose harmless bacteria to antibiotic disks and measure the clear zones where growth stops, then compare strains or repeated exposures. If you can't get lab access, you can analyze public antibiotic resistance surveillance data and look for patterns across regions or years.


Best for: You if you're interested in medicine and global health.


Requirements: A supervised lab for the wet version, or a spreadsheet and public data for the data version.


4. The Role of Enzymes in Food Digestion

You test how temperature or pH changes the speed at which an enzyme such as amylase breaks down starch. A color change with iodine shows you how much starch is left. You can connect your results to what happens in your own digestive system.


Best for: You if you want a first biochemistry project with clear, visible results.


Requirements: A thermometer, a timer, starch and enzyme samples, and a teacher or adult to supervise the iodine.


5. The Impact of Soil Composition on Plant Growth

You grow the same seeds in sand, clay, and loam and track height, leaf count, and plant mass over several weeks. You can also compare how much water each soil holds, or add fertilizer to one group.


Best for: You if you like slow, steady experiments and want to watch results grow on a windowsill.


Requirements: Seeds, pots, a ruler, a kitchen scale, and a few weeks of daily checks.


6. How Sleep Relates to Memory in Teenagers

You review published studies, or run a small anonymous survey, to see how hours of sleep line up with scores on a short recall test. You then look at what the data can and can't tell you, since a link between two things doesn't prove one causes the other.


Best for: You if you're interested in the brain and comfortable with basic statistics.


Requirements: Permission from a teacher or parent, consent from every volunteer, no names attached to the data, and a spreadsheet.


7. Spaced Practice vs. Cramming

You compare how well volunteers remember a word list after one long study session versus the same time split across several days. Then you graph the results and test whether the difference is bigger than chance.


Best for: You if you want a psychology project with quick, measurable results.


Requirements: Volunteers who give consent, flashcards or a word list, and a spreadsheet for simple statistics.


8. Heart Rate Recovery After Exercise

You measure how fast pulse returns to normal after a standard step test, then compare groups or conditions, such as people who exercise often and people who don't. You can also test whether time of day changes the results.


Best for: You if you like human biology and collecting your own numbers.


Requirements: A stopwatch, volunteers with permission and consent, and a safe activity. Skip anyone who has a health condition that makes exercise risky.


9. Plant Growth Under Different Colors of Light

You grow seedlings under red, blue, and white LED bulbs, keeping everything else the same, and measure height and leaf area every few days. The results can show you which light the plants use best.


Best for: You if you want a clean experiment with one variable and a clear answer.


Requirements: LED bulbs in different colors, a timer, seedlings, a ruler, and a dark corner you can control.


Environmental science and Earth

These topics send you outside or have you work with real environmental data. For programs that go further, see our list of environmental science research opportunities.


10. Investigating Microplastic Pollution in Local Water

You collect water from a lake, river, or tap, filter it, and count tiny plastic particles under a microscope, sorting them by size and shape. Compare sites or seasons to look for patterns, and think about where the plastic might come from.


Best for: You if you want a project that connects to a real environmental problem.


Requirements: A microscope, filter paper, clean technique so fibers from your clothes don't contaminate samples, and an adult with you when you sample near water.


11. Designing a Low-Cost Water Filtration System

You build filters from sand, charcoal, and gravel, then measure how cloudy the water is before and after. Compare designs to find the most effective one, and work out what each costs.


Best for: You if you like building things and solving practical problems.


Requirements: Clear containers, a way to measure cloudiness (a turbidity tube or a phone light method), and a firm rule that nobody drinks the filtered water.


12. Air Quality Analysis Using Low-Cost Sensors

You use an affordable particle sensor to log PM2.5 over days or weeks, then compare readings with traffic, weather, or time of day. You can also compare indoor and outdoor air and graph what you find.


Best for: You if you enjoy electronics and data together.


Requirements: A low-cost particle sensor and logger, a spreadsheet, and a note in your write-up that cheap sensors are less precise than professional ones.


13. Renewable Energy Efficiency in Urban vs. Rural Settings

You measure the output of a small solar panel in two different places, such as a city rooftop and an open field, at the same times of day. Then you look at how shading, temperature, and air quality might explain the difference.


Best for: You if you're interested in clean energy and fieldwork.


Requirements: A small solar panel, a multimeter, a light meter or phone app, and access to two sites.


14. How Urban Green Spaces Support Biodiversity

You compare a park with a roadside strip, counting the plants, birds, or insects you find in each over several visits. Then you look at how size, plant cover, and human activity might explain the differences.


Best for: You if you like nature and don't mind early mornings outside.


Requirements: A field guide or identification app, a notebook, a spreadsheet, and permission to visit your sites.


15. Backyard Bird Counts and the Weather

You count bird species at the same spot and time each day, then compare your counts with temperature, wind, or rain. After a few weeks you can graph the patterns and look for a relationship.


Best for: You if you want a low-cost project that fits into your daily routine.


Requirements: A field guide, binoculars if you have them, a consistent schedule, and a weather log.


16. Rainwater pH Across Different Storms

You collect rain in clean containers and measure its pH with strips or a meter, then compare storms, wind directions, or locations. Normal rain is slightly acidic, so your question is how much yours varies and why.


Best for: You if you like chemistry and want a project that depends on the weather.


Requirements: Clean collection containers, pH strips or a meter, and a rain log.


17. Mapping Urban Heat Islands

You place thermometers or temperature loggers at shaded and paved spots around your neighborhood and record readings at the same times. Then you map where it runs hottest and look at what those places have in common.


Best for: You if you want to connect temperature data to maps and city planning.


Requirements: Several thermometers or loggers, a consistent schedule, and a simple map of your sites.


18. Soil Erosion on Different Slopes

You tilt trays of soil at different angles, pour the same amount of water over each, and collect the sediment that washes away. Then you repeat with grass or mulch on top to see how ground cover changes the result.


Best for: You if you like building a test setup and measuring something physical.


Requirements: Trays, soil, a watering can, a sieve, and a kitchen scale.


Physics and engineering

These topics are about measuring, building, and testing.


19. The Physics of Projectile Motion in Sports

You film basketball shots or soccer kicks, track the ball frame by frame in free video analysis software, and calculate angle, speed, and range. Then you compare your numbers with what the equations predict and look at what spin or air resistance changes.


Best for: You if you love sports and want to see the physics behind them.


Requirements: A phone on a tripod, a free video analysis tool, and basic trigonometry and kinematics.


20. The Effect of Light Intensity on Solar Cell Efficiency

You shine a lamp on a small solar cell from different distances and record voltage and current at each one. Plot the results to see where output levels off, and try changing the angle or temperature too.


Best for: You if you want a clean physics experiment with clear graphs.


Requirements: A small solar cell, a multimeter, a lamp, a ruler, and some unit conversions.


21. The Effect of Temperature on Liquid Viscosity

You time how fast water, oil, and glycerin flow or let a small ball sink through them at different temperatures. Then you plot viscosity against temperature and compare how each liquid responds.


Best for: You if you like physical chemistry and measuring small differences.


Requirements: A tall clear container, a stopwatch, a thermometer, and an adult to help with any heated water.


22. Air Resistance and Falling Coffee Filters

You drop stacks of coffee filters from a fixed height and time the fall, then see how weight changes the speed they reach. Graphing your results shows how air resistance pushes back harder as objects speed up.


Best for: You if you want a simple physics experiment you can run in an afternoon.


Requirements: Coffee filters, a stopwatch or phone video, a safe measured drop height such as a stairwell with an adult, and a spreadsheet.


23. Truss Design and Bridge Strength

You build small bridges from craft sticks using different truss patterns, add weight until each one breaks, and compare strength to the amount of material used. The results show you which shapes carry load best.


Best for: You if you like engineering and don't mind a few broken bridges.


Requirements: Craft sticks, glue, weights, safety goggles, and a consistent way to load each bridge.


24. Sound Insulation Materials

You place a speaker playing a steady tone inside a box lined with foam, cardboard, or fabric and measure how much quieter it gets outside. Compare the materials and think about how thickness changes the result.


Best for: You if you're interested in acoustics and building design.


Requirements: A sound level meter app (check it against a known sound), a speaker, boxes, and test materials.


25. Wind Turbine Blade Design

You make small turbine blades with different shapes, angles, or numbers of blades, spin them in front of a fan, and measure the voltage a small motor produces. Then you find which design gives the most power.


Best for: You if you want to design, test, and improve something over several rounds.


Requirements: A small DC motor, a fan, a multimeter, and craft materials.


Computer science, AI, and data

You can do all of these on a laptop. Our guide to online research programs for high school students lists free and paid options if you want a mentor for a project like these.


26. Machine Learning for Predicting Student Performance

You train a model in Python with scikit-learn to predict academic outcomes from a public dataset, using factors like study time or attendance. Test several algorithms, such as regression and decision trees, and compare how accurate each one is.


Best for: You if you want your first data science project.


Requirements: Basic Python (or Excel), a public education dataset with no personal information, and an understanding of train and test splits.


27. AI-Based Image Recognition for Plant Disease Detection

You train a model on leaf photos to tell healthy plants from diseased ones, using a browser tool like Teachable Machine or a library like TensorFlow. Then you test it on new images and see how photo quality or the number of training pictures changes accuracy.


Best for: You if you like AI and plants.


Requirements: A public leaf image dataset, free browser tools, and optional beginner Python.


28. Exploring Neural Networks Through Game Strategy

You build a small neural network that learns tic-tac-toe or a similar game by playing thousands of rounds against itself. Then you study how it improves, where it still loses, and how changing the network's size affects learning.


Best for: You if you already code and want to see what's going on inside a machine learning model.


Requirements: Python, comfort with loops and functions, and some patience for debugging.


29. Sentiment Analysis of Public Text

You run a sentiment model on public product reviews or news headlines, then label a sample by hand and compare. The gap between the model and your labels shows you where it gets confused by sarcasm or slang.


Best for: You if you like language and want to see how computers read it.


Requirements: Python, a public text dataset, and the patience to label about 100 samples yourself.


30. Predicting Local Temperature With Simple Models

You download several years of weather records for your area and compare a moving average with a linear regression to predict next week's temperature. Then you measure which one makes smaller errors.


Best for: You if you want a forecasting project with data you can check yourself.


Requirements: Public weather records, a spreadsheet or Python, and basic statistics.


31. Checking a Public Dataset for Bias

You train a simple model on a public dataset that includes group labels, then check whether its accuracy differs between groups. You write up what you find and what might cause the gap.


Best for: You if you care about fairness in technology.


Requirements: Python with pandas, a public dataset with group labels, and careful reading about how the data was collected.


32. Analyzing Public Transit Delays

You take open data from your local transit agency and look at how delays change by time of day, route, or weather. Then you chart your results and suggest one change that could help riders.


Best for: You if you ride the bus or train and wonder why it's late.


Requirements: A spreadsheet or Python, and an open data portal from your local agency.


33. How Image Compression Affects Quality

You save the same photos at different JPEG quality settings and compare file size with how different the picture looks, using a simple pixel-difference score. Then you find where extra quality stops being worth the bigger file.


Best for: You if you like the math behind everyday technology.


Requirements: A set of photos, Python with the Pillow library or an image editor, and a way to measure difference between two images.


Chemistry, math, and materials

34. Making Biodegradable Plastic From Starch

You cook starch films with different amounts of glycerin, then test how far each one stretches and how fast it breaks down in soil. The results show you how a recipe change shifts the material's properties.


Best for: You if you want chemistry with a sustainability angle.


Requirements: Cornstarch, glycerin, vinegar, a kitchen scale, an adult for the stove, and soil for breakdown tests.


35. Growing Crystals Under Different Conditions

You grow crystals from salt or alum solutions at different concentrations or temperatures and measure how big they get and how fast. Then you look at how the setup changes the shape and size.


Best for: You if you want a patient, visual experiment.


Requirements: Salt or alum, jars, string, hot water with adult help, and a place to leave jars undisturbed. Keep alum away from young kids and pets.


36. Testing Benford's Law on Real Datasets

Benford's law says the first digit of many real-world numbers is more often 1 than 9. You check whether it holds for datasets like city populations or river lengths, and find places where it fails.


Best for: You if you like math puzzles and surprising patterns.


Requirements: A public dataset, a spreadsheet, and an intro to logarithms and probability.


37. Monte Carlo Simulations

You use random numbers to estimate pi or the odds in a card game, then see how the answer settles as you run more trials. Graphing the convergence shows you why simulation is so useful.


Best for: You if you like probability and coding.


Requirements: Python or a spreadsheet's random function, and basic probability.


Where can you take your project next?

Once you finish, you have a few ways to share it. The table compares four of them.


Option

Who can take part

Format

What you finish with

High school students, no research experience needed

Fully online, one-on-one with a PhD fellow or postdoc, 12 weeks in the Individual Research Program

A college-level research paper

Middle and high school students, guided by a teacher or advisor

Submit original research for peer review

Possible publication in an open-access journal

Students at all levels

Collect or analyze environmental data and report findings

A research report shared with other students and scientists

High school students who advance through affiliated science fairs

Science fair competition

A project presentation judged for awards


If you finish a project in the spring, you could enter a science fair, send it to a journal for review, or take it to a mentor for a longer paper. Our guide to publishing a research paper in high school walks through the steps.


Want to research your own STEM question?

These topics are starting points. If you'd like to go deeper on one, the Lumiere Research Scholar Program pairs you one-on-one with a PhD fellow or postdoc. In the 12-week Individual Research Program, you meet your mentor 9 times, work with a writing coach for 2 sessions, and finish with a college-level research paper. It's fully online; you don't need research experience, and need-based financial aid is available. Completing a program can also make you eligible for 3 post-baccalaureate credits through UC San Diego Extended Studies.


What you might still be wondering about these STEM projects

Can I test my ideas on classmates or family?

Sometimes. The sleep, memory, and heart rate topics involve people, so you need permission from a teacher or parent, consent from each volunteer, and no names attached to the data. The data topics skip that step because no people are involved.


Do I need to know how to code?

Not for most of them. Many topics use a spreadsheet or simple tools, and the plant disease idea can start with a free browser tool. Coding helps most with the machine learning, neural network, and text analysis topics.


Can a project like these get published?

Yes, some journals accept original work from high school researchers, and most expect a teacher or mentor to guide the project. Expect revisions or a rejection on the first try, and read our guide to publishing a research paper in high school before you submit.


Do I need to live near a university to get help with a project?

No. You can start with a teacher at your school or an online guide. If your school has no research program, the Lumiere Research Scholar Program is fully online and pairs you one-on-one with a PhD fellow or postdoc, with need-based aid available.


Resources 

External resources


Resources from Lumiere

a transparent logo of Lumiere education
  • Facebook
  • Instagram
  • Twitter
  • LinkedIn

+1 ​‪(573) 279-4102‬

​

United States of America: 300 Delaware Avenue, Suite 210 #207, Wilmington, DE 19801, USA

​

United Kingdom: Camburgh House, 27 New Dover Road, Canterbury, Kent, United Kingdom CT1 3DN

We are an organization founded by Harvard and Oxford PhDs with the aim to provide high school students around the world access to research opportunities with top global scholars.

​

​

​

©2024 by Lumiere Education.

bottom of page