Writing a Science Fair Procedure That Actually Works

A procedure is supposed to let someone else repeat your experiment exactly. That sounds simple but it is the part where most science fair projects fall apart. Judges will read your procedure and try to picture themselves doing it. If they cannot, you are going to lose points whether your actual results were good or not. The most common mistake I see is under-specification. A student will write something like "mix the two chemicals" and then wonder why their replication failed. Mix them how? At what ratio? In what container? For how long? Those missing details are the difference between a procedure that passes review and one that gets sent back for revision.

Science Fair Procedure Examples should be written so that someone with basic lab training could execute them without asking you a single question. That means every measurable quantity has a number attached to it, and every tool has a specification when it matters.

Here is the structure I use when I write procedures for my own work. It applies directly to science fair projects.

Step-by-step procedure format

Start with a title and date. This sounds trivial but judges sometimes sort through dozens of project boards and loose sheets of paper. A clear header prevents confusion. Then list your materials. Every item that physically touches your samples or data goes on that list. Include quantities and specifications. A graduated cylinder is not enough information if you used a 10 mL cylinder for one measurement and a 100 mL cylinder for another. Specify which one and when. Write your steps as numbered items. Use sub-bullets for anything that branches into conditional actions. Do not write paragraphs inside the procedure section. Paragraphs hide important details inside blocks of text. Judges skim procedures looking for specific things. Make it easy for them to find those things. Each step should answer what you do, how much you do it with, and how you know when to move to the next step. I once had a student who wrote "heat until dissolved." There was no temperature, no time specification, no visual cue beyond a subjective "dissolved." If someone reheated his experiment later, they would have had no idea what he actually did. The corrected version specified "Heat to 75 degrees Celsius on a hot plate set to level 4. Stir continuously with a glass rod. Record the time when the last visible solid particle disappears." That is the level of detail most judges expect.

I ran into a specific edge case once that I still think about. A student was testing how different concentrations of salt water affect seed germination. His procedure said "soak seeds overnight in salt water solution." Overnight means different things to different people. More critically, he had prepared three separate salt concentrations but never specified how he labeled which beaker was which concentration. When a judge asked him to explain how someone else would replicate his setup, he realized he had written nothing about the labeling method. He had relied on visual cues from the marker pen he used on the beakers, but he never included that in the procedure. We added a line specifying that each beaker was labeled with the concentration in millimoles per liter using a permanent black marker applied to the lower third of the glass. That small addition resolved the entire ambiguity and the judge moved on.

The bigger issue behind that story is what I call the labeling gap. Most students treat labeling as a practical detail that does not belong in the procedure. It does belong there. If you have multiple test conditions, your procedure must explain how each condition is identified and distinguished during the experiment. Without that, the procedure is incomplete even if the physical steps are described accurately.

Controlled variables and constants

This is where procedures get nuanced. You need to separate what you intentionally change from what you intentionally hold constant. A standard procedure section does not always make this distinction obvious. I recommend adding a short subsection right after your materials list called "Variables" that explicitly states your independent variable, your dependent variable, and your constants. Independent variable is the thing you manipulate. Dependent variable is the thing you measure. Constants are everything else you keep the same. Students frequently conflate constants with controls. A control is a specific condition, usually a baseline with no treatment or a standard comparison. Constants are parameters you keep fixed across all conditions. The difference matters when a judge asks you to justify your experimental design. I have seen students lose points because their constants section listed environmental factors they did not actually measure or monitor. If you claim room temperature was a constant, you should have a way to verify it stayed within a reasonable range. A procedure that states "maintained at room temperature" without any mention of monitoring or acceptable variation is a red flag. Write "Room temperature was recorded at the start and end of each trial using a digital thermometer. All trials occurred between 21 and 23 degrees Celsius." Now it is defensible.

Replication and trial counts

Your procedure should state how many trials you plan to run and why that number is sufficient. Three trials is the minimum I accept for most middle school and high school projects. Fewer than that and your results are vulnerable to random variation. More than five trials without justification starts to look like padding rather than good science. When I write procedures now, I include a line about replication strategy. Something like "Three trials per condition were conducted on separate days to account for daily environmental variation. A fourth trial was prepared as a contingency but was only used if any of the initial three trials showed anomalous results due to equipment malfunction." That tells a judge you thought about reproducibility and error handling without inflating your trial count artificially. I also recommend documenting modifications between trials. If you changed a step after the first trial based on what you observed, note that change in the procedure itself. Write it as a revised step rather than hiding it in your conclusion or discussion. Judges appreciate transparency about iteration. They do not appreciate the implication that your first attempt was perfect.

Common pitfalls to avoid

One pitfall is mixing procedure and results in the same section. Keep observations and data collection out of the procedural steps unless the observation itself is part of the action you are describing. "Record the temperature at five minute intervals" belongs in the procedure. "The temperature rose to 42 degrees Celsius" belongs in your results section. Another pitfall is using ambiguous time references. "Wait five minutes" is not the same as "Allow the mixture to sit for five minutes at ambient temperature before proceeding." The second version tells the reader exactly what waiting entails. The first version leaves open whether stirring occurs during that time, whether the container is covered, and whether ambient conditions matter. A third pitfall is omitting safety precautions within the procedure itself. Safety goes in the procedure, not in a separate disclaimer section that judges might skip. If you handle hot liquids, sharp tools, or chemical reagents, write the safety step inline. "Wear nitrile gloves when handling the hydrogen peroxide solution. Work in a well-ventilated area. Dispose of waste in the designated chemical disposal container."

What good Science Fair Procedure Examples share

They are precise without being verbose. They specify quantities, tools, timing, and conditions. They separate variables clearly. They account for replication and error handling. They include inline safety notes. They anticipate how a reviewer will try to break them and close those gaps preemptively. Writing a complete procedure typically takes about ten to fifteen minutes for a standard project. The ones that get flagged for revisions usually take an additional twenty to thirty minutes of revision because the original draft was too vague on critical parameters. Spending those extra minutes upfront saves you from scrambling during the judging period. There are limits to how much detail you can include. Some experiments involve tacit knowledge that is hard to write down, like the feel of when a solution has reached the right viscosity. In those cases, include your best approximation and note where judgment is required. A judge would rather see an honest acknowledgment than a procedure that pretends every decision is purely mechanical. If your experiment is highly variable by nature, consider running a pilot procedure first. Write the full procedure, then execute it once as a test run. You will almost certainly find steps that are ambiguous, equipment you did not account for, or timing constraints that do not work as expected. Fix those issues in the procedure document before you start your actual trials. This usually catches two or three problems per project that would otherwise show up during judging.