Getting Started
What Do Science Fair Judges Look For? A Scoring Guide for High School Projects
Judges score eight things: your question, background research, hypothesis, method, data, conclusions, understanding and communication. Here is what each one looks like in a strong project, plus a practice rubric and the interview questions to expect.
Sir Luther Center - Research & Innovation Team - 11 min read
How Science Fair Judging Usually Works
Every fair runs its own process, so the first step is always the same: read your own fair's rules and rubric. A school fair, a regional fair and a national competition each score differently, and their official documents are the only ones that count on the day.
That said, most fairs follow the same broad shape, and knowing the shape tells you where to put your effort.
Judging usually happens in three stages. First come the written materials: the report, research plan or project summary you submit before the fair. Second is the display review, where judges walk the floor and read boards, sometimes before students are allowed to speak. Third is the interview, a few minutes of questions at your board.
Students tend to over-invest in the board and under-prepare for the interview. The board gets a judge to stop; the interview is where the scoring actually happens. A judge who has read hundreds of boards can tell within two questions whether the student in front of them did the work or watched someone else do it.
Who are the judges? Usually a mix of teachers, university researchers, engineers and industry professionals who volunteer for the day. Most are not specialists in your exact topic, which is why a clear explanation matters so much. A judge who cannot follow your project cannot score it highly, no matter how good the science is.
One more thing worth knowing: judges typically score each project against the rubric, not against each other, and then compare scores at the end. Your job is not to look better than the student next to you. It is to tick every box on the sheet, which is good news, because the boxes are knowable.
The 8 Things Judges Score
Science fair judging criteria vary between fairs, but the same eight areas appear again and again, plus one pass-or-fail check. Here is what each looks like in a strong project, and the mistake that quietly costs marks.
| What judges look for | What it looks like in a strong project | Common mistake |
|---|---|---|
| Research question and originality | One narrow question a stranger could repeat back, with a personal angle | A topic so broad no experiment could answer it |
| Background research | A mini literature review: what earlier studies found and what gap remains | Three paragraphs copied from Wikipedia |
| Hypothesis and variables | A testable prediction, with independent, dependent and controlled variables named | A "hypothesis" that is really just a topic |
| Method and controls | A procedure another student could repeat, with a control group and enough repetitions | Changing two things at once, or running the test once |
| Data and analysis | Tables with units, a graph you made yourself, averages and honest notes on odd results | A single photograph and the word "results" |
| Conclusions and limits | An answer tied to the data, plus what the project cannot prove | Claiming the results "prove" something they cannot |
| Understanding (the interview) | Calm, specific answers about your own numbers and choices | Memorised speech that collapses at the first follow-up |
| Display and communication | A board a stranger can follow in two minutes, in a logical order | Crowded text, tiny fonts and decorations instead of content |
| Safety and ethics (pass-or-fail) | Rules on human participants, animals, chemicals and hazards followed and documented | Assuming nobody will check |
A few of these deserve a closer look.
The question. Judges read your question first, and many form their initial impression from it alone. "Does music affect plants?" is a topic. "Does exposure to classical versus rock music change the germination rate of basil seeds?" is a question. If you are still choosing, our list of science fair project ideas for high school students shows what the strong version of more than 60 projects looks like, and the Topic Finder can help you narrow a vague interest into something testable.
Background research. This is the most skipped step, and judges notice. Strong projects read what others have already done, like a mini literature review, and use it to justify the question. Our step-by-step guide to writing a literature review shows how to compare sources instead of just summarising them.
Data and analysis. Judges want to see that you collected enough data to mean something and that you actually looked at it. A table with units, a graph you made yourself, an average across repeated runs: these signal care. So does honesty about the messy parts. A student who says "this run went wrong because the room was cold, so I repeated it" sounds like a researcher. A student with suspiciously perfect data sounds like a risk.
The interview. This is where understanding is scored. Judges are not trying to catch you out; they are checking that the work is yours. If you can explain why you chose your method, what your weirdest data point means and what you would do differently, you will score well here.
Display and communication. A good board is not a decorated board. It is one a stranger can follow in two minutes: question, method, data, conclusion, in an order that makes sense. Use short labels, readable fonts and graphs instead of paragraphs. Everything on the board should earn its place; if a photo or border adds nothing, cut it.
Safety and ethics. Every fair has its own rules on human participants, animals, chemicals and hazardous materials, and you must follow them. These are usually pass-or-fail: a project that breaks them can be removed regardless of its science. Get the approvals your fair requires before you start, not after.
A Practice Rubric You Can Score Yourself With
This is a practice rubric we wrote for self-assessment. It is not any fair's official scoring sheet, and your fair's own rubric always takes priority. Use it to find weak spots while there is still time to fix them.
Score each row from 0 to 4: 0 means missing, 2 means present but weak, 4 means a judge would struggle to criticise it.
| Criterion | 0 | 2 | 4 |
|---|---|---|---|
| Research question | No clear question | Question exists but is broad | Narrow, testable, repeated back easily |
| Background research | None | A few sources, summarised | Sources compared, gap identified |
| Hypothesis and variables | Missing | Hypothesis without named variables | Prediction plus all variables named |
| Method and controls | Unrepeatable | Mostly clear, gaps in controls | Repeatable, controlled, repeated runs |
| Data and analysis | No usable data | Numbers present, thin analysis | Tables, graphs, averages, anomalies noted |
| Conclusions and limits | Unsupported claims | Conclusion exists, limits ignored | Answer tied to data, limits stated |
| Interview readiness | Cannot explain the work | Explains what, not why | Explains choices, data and next steps |
| Display and communication | Confusing | Readable with effort | Followable in two minutes |
Add up your score out of 32. Below 20, fix the science before touching the board. Between 20 and 27, your weakest row tells you where the next hour of work goes. Above 28, spend your remaining time rehearsing the interview, because that is where close contests are decided.
What Separates a Winning Project From a Good One
Good projects are common at every fair. Winning ones tend to share a handful of specific traits, and none of them require expensive equipment. If you are wondering how to win a science fair, this list is the honest answer: there is no trick, only a set of habits that judges consistently reward.
- The student owns one decision. A winning project has a moment where the student made a real choice: an unusual variable, a method they adapted, a problem they solved. Judges ask about it, and the answer is memorable.
- The data has texture. Winners report the odd result, the failed run, the day the thermometer broke, and explain what they did about it. Perfect data with no story reads as borrowed.
- The conclusion is smaller than the project. It sounds backwards, but students who claim less score more. "In my samples, under my conditions" beats "this proves" every time.
- The limits are named before the judge asks. Saying "my sample was small and from one school" first turns a weakness into evidence of understanding.
- The next experiment is ready. When a judge asks "what would you do next?", a winning student has a specific answer, not a shrug.
- The project connects to something real. A sentence about why the question matters outside the fair, kept honest and modest, lifts the whole board.
Questions Judges Often Ask
How are science fair projects judged in the interview? Mostly through questions like these. Prepare an honest answer for each.
- "Why did you choose this project?" Give the real reason, even if it is small. Genuine curiosity is obvious and scored well.
- "What was your control, and why?" Name it and explain what would go wrong without it.
- "What surprised you in your data?" Have a specific data point ready. "Nothing" suggests you did not look closely.
- "What would you do differently?" Pick one real improvement: more repetitions, a better measurement, a tighter variable.
- "How does this connect to earlier research?" This is where your background reading pays off. Name one study or finding you built on.
- "What are the limits of your conclusion?" State what your data cannot show. This question separates strong projects from rehearsed ones.
- "Did anyone help you?" Answer honestly. Help from a parent or teacher is normal; pretending otherwise is the problem.
- "What would the next experiment be?" Describe the follow-up you would actually run.
- "Why does this matter?" One modest, honest sentence about the real-world angle is enough.
Common Mistakes
These six show up at every fair, and every one of them is fixable if you catch it early enough.
- Choosing a topic instead of a question. "Volcanoes" is not testable. "Does vinegar concentration change eruption height?" is.
- Running the experiment once. One run is luck. Three or more runs per condition is evidence.
- Changing two variables at once. You will never know which one caused the result.
- Hiding the weird data. Odd results are where the interesting science lives. Report them.
- Overclaiming the conclusion. Match your claims to what your data actually shows, no more.
- Memorising a speech for the interview. Judges ask follow-ups. Know your project instead of a script.
Your Checklist for the Week Before Judging
- Reread your fair's rules and rubric, including safety and ethics requirements.
- Score your project with the practice rubric above and fix the weakest row.
- Check every number on your board against your original data table.
- Rehearse a two-minute explanation, then practise answering the nine questions above out loud.
- Prepare one sentence on your background reading and one on your project's limits.
- Print any forms your fair requires, and pack spare supplies for display day.
One last thing: sleep the night before. A tired student gives flat answers, and the interview is too big a share of the score to spend it yawning. The project is already done by then; the only job left is to show the judges the work you actually did.
A Worked Example: Weak Version vs. Strong Version
This is an invented example to show how the same idea scores differently. The project is plant growth, a fair staple.
Same topic, same kitchen equipment. The difference is a narrower question, a fair test, real data and honest limits. That is the whole game.
From Science Fair Project to Research Paper
A strong science fair project is already a small piece of research: a clear question, background reading, a fair method, honest data and a written report. What separates it from a publishable paper is mostly depth. A paper needs a fuller literature review, a question nobody has quite answered yet, more careful writing and the scrutiny of peer review, where outside experts check the work before it is published.
Some students take a strong project further as they get older, and the path is well worn. The report you wrote for the fair becomes the draft; the judges' questions become the reviewer comments you prepare for. Our guides on how to write a research paper as a high school student and how to publish research as a high school student walk through both stages, and you can browse finished student papers to see what the end result looks like.
If you want structured help making that jump, our explainer on what a research mentorship program is describes how mentor support works, and the Research Publication Fellowship is one option for students in grades 8-12, with Harvard student mentors guiding one project from question to submission. Younger students can start smaller: our science fair project ideas for middle school students builds the same habits with simpler tests.
Frequently Asked Questions
- Do judges care more about results or process?
- Process, almost always. A careful experiment with a surprising or null result scores better than a flashy result from a sloppy method. Judges want to see that you asked a real question and tested it fairly, because that is what research actually is.
- Do I need to have a new idea?
- No. Very few student projects are completely original, and judges know that. What they look for is a personal angle: a new variable, a new context, a better method or a question you genuinely care about. Repeating a known experiment with your own careful twist is a respected approach.
- What if my experiment fails?
- A failed experiment is not a failed project. If your hypothesis was wrong or your results were inconclusive, say so, explain what the data does show and describe what you would change next time. Judges reward honest analysis of a null result far more than a claimed success they cannot verify.
- How much do the display and speaking matter?
- They matter, but they cannot rescue weak science. A clear board and a confident two-minute explanation get judges to engage with your work, and the interview is where they confirm you did it. Think of presentation as the delivery system for the research, not a substitute for it.
- How are high school projects judged compared with middle school?
- The core criteria are the same, but the bar rises. High school judges expect real background reading, more repetitions, proper controls and a discussion of limitations. A middle school project can win with a clean, simple test; a high school project needs to show you understand where your work sits among earlier studies.
- Can I use ChatGPT for my science fair project?
- Check your fair's AI rules first, since they vary and some require disclosure. Brainstorming questions, asking what variables to control and checking grammar are generally fine. Inventing data, generating fake sources or letting it write your report is not, and the interview will expose anything you did not actually do.