# Science Fair Project Proposal Template By the end of this guide you will be able to fill out a complete science fair project proposal, from the research question through the [data management](/blog/guides/data-management-basics-principles-processes-and-best-practices) plan, using a copy-and-paste template. You will also know how reviewers read proposals, why two judges can score the same project differently, and how to keep a [lab notebook](/knowledge/molecular-biology/lab-notebook) that makes your final report easy to write. To follow along you need a word processor or a plain-text editor, a spreadsheet program such as Google Sheets or Excel, and a notebook or a printable lab notebook template PDF for recording data. Nothing needs to be installed beyond that. A printed copy of the template below plus a pencil is enough to get started. ## What a Proposal Is and Why Judges Read One A science fair project proposal is a short written plan for a project you have not done yet. It states what you will test, why it matters, how you will test it, what materials you need, and what you will do with the data. Most fairs require the proposal before you start because the safety review, the adult sponsor sign-off, and the judging schedule all depend on it. Professional scientists do the same thing on a larger scale. Research funding almost everywhere runs on written proposals, and the way those proposals are judged shapes what science gets done. One research group has even proposed replacing traditional proposal review with a system where scientists donate portions of their funding directly to colleagues they trust, which they argue would reduce some of the bias and overhead in conventional review [1]. You are not competing for grant money, but the core skill is identical. You are asking a skeptical reader to believe that a specific, planned piece of work is worth doing and that you can do it. That skepticism is real even among experts. In a study of 248 research grant proposals submitted simultaneously to two similar funding agencies using the same scoring system, agreement between the agencies was only fair beyond chance, with a [Cohen's kappa](/blog/research-skills/cohens-kappa-how-to-calculate-and-interpret) of 0.29 [2]. Cohen's kappa is a statistic that measures agreement between raters after correcting for agreement you would expect by luck. Values near zero mean the raters barely agreed at all, and values near one mean near-perfect agreement. When the analysis looked only at proposals that were either "clearly fundable" or "clearly not fundable," agreement rose to a kappa of 0.444, still only moderate [2]. The lesson for you is straightforward. Clear, concrete proposals land in the "clearly fundable" category, where reviewers agree. Vague proposals land in the middle band, where reviewers disagree with each other and with your intentions. ## The Template Format Here is a complete science fair project proposal template format you can copy into a document. Most school and regional fairs accept this structure or something close to it. Check your fair's own rules first, because some require specific forms or safety certifications on top of this. ``` SCIENCE FAIR PROJECT PROPOSAL 1. Project Title 2. Student Name(s) and Grade Level(s) 3. Adult Sponsor / Teacher 4. Research Question 5. Background Research (150-250 words) 6. Hypothesis 7. Variables - Independent variable: - Dependent variable: - Controlled variables: - Control group: 8. Materials and Cost Estimate 9. Procedure (numbered steps) 10. Data Collection Plan - What you will record: - How often: - Where you will record it (lab notebook): 11. Data Management Plan - How data will be stored and backed up: - How you will label files and records: - Whether data will be shared: 12. Sample Size and Analysis Plan 13. Safety Considerations 14. Timeline 15. Acknowledgment of Help ``` Each numbered section answers one question a reviewer will ask. The table below summarizes the purpose of each section and the most common mistake in it. | Section | What the reviewer wants to know | Most common mistake | |--|--|--| | Research Question | Is this testable and specific? | A topic instead of a question, such as "plants and salt" | | Background Research | Do you know what is already known? | Citing no sources or only general websites | | Hypothesis | Can it be proven wrong? | "I think my project will work" | | Variables | Is the experiment controlled? | More than one thing changing at a time | | Materials | Can someone repeat this exactly? | Vague entries such as "some soil" | | Procedure | Can a stranger follow it step by step? | Missing quantities, times, and temperatures | | Data Plan | Will you actually capture usable data? | No plan for recording before the experiment starts | | Data Management | Will the data still make sense in three months? | Files named "data1" and "data_final2" | | Sample Size | Will the result mean anything? | Testing one plant per condition | | Safety | Did you think about risks? | Leaving the section blank | | Timeline | Is the project feasible before the fair? | No calendar dates at all | | Acknowledgment | Who actually helped, and how? | Not mentioning help at all | ## Filling It In: A Worked Example This science fair project proposal template example uses a simple, safe, repeatable experiment on seed germination. Follow the numbered steps and the filled template below. ### Step 1. Write the research question and hypothesis Start with a question you can answer with a measurement. A working pattern is "How does changing [one thing] affect [one measured outcome] in [a specific organism or object]?" Our example question: How does the concentration of salt in watering solution affect the germination rate of bush bean seeds? The hypothesis predicts a direction and can be proven wrong: Bush bean seeds watered with salt solutions at 0.5, 1.0, and 2.0 percent salt by weight will show lower germination percentage after 7 days than seeds watered with plain water, and germination will fall as salt concentration rises. Research on how researchers formulate questions supports this kind of structured approach. A consensus study using the nominal group technique, a method where experts generate and rank ideas in structured rounds, identified established workflows for research question definition and recommended combining the linear, deductive style common in health research with the more iterative style common in data science [3]. For a science fair proposal, the practical version is simple. Write the question first, make it measurable, then write the hypothesis as a specific prediction that the data could contradict. ### Step 2. Define the variables - Independent variable, the thing you change: salt concentration of the watering solution, at 0, 0.5, 1.0, and 2.0 percent by weight. - Dependent variable, the thing you measure: germination percentage at 7 days, defined as the fraction of seeds with a visible root at least 2 millimeters long. - Controlled variables, the things you hold constant: seed variety and source, 40 seeds per group, 50 milliliters of solution per pot per day, the same potting mix, the same pot size, the same location, and room temperature. - Control group: the 0 percent salt group. ### Step 3. List materials with exact quantities - 160 bush bean seeds from one packet - 8 small pots, 10 centimeters diameter - 8 liters of standard potting mix - Table salt, measuring spoon, and a kitchen scale that reads to 1 gram - Graduated cylinder or measuring cup - Tap water - Ruler with millimeter markings - Lab notebook or printed lab notebook template ### Step 4. Write the procedure as numbered steps someone else could follow ``` 1. Prepare 2 L of 2.0% salt solution: dissolve 40 g NaCl in 2,000 mL tap water. 2. Prepare 2 L of 1.0% salt solution: dissolve 20 g NaCl in 2,000 mL tap water. 3. Prepare 2 L of 0.5% salt solution: dissolve 10 g NaCl in 2,000 mL tap water. 4. Prepare 2 L of 0% solution: plain tap water. 5. Fill 8 pots with equal volumes of potting mix. 6. Plant 20 seeds per pot, 2 cm deep. 7. Assign pots to groups: 2 pots per treatment, randomized positions on the same windowsill. 8. Water each pot with 50 mL of its assigned solution daily at 7 AM. 9. Count germinated seeds at 7 days. A seed counts as germinated when the root is visible and at least 2 mm long. 10. Record counts in the lab notebook the same day. ``` ### Step 5. Plan the sample size and the analysis ![Worked calculation: 36 of 40 seeds gives 90.0 percent, 12 of 40 gives 30.0 percent, a difference of 60.0 percentage points.](/article-images/illustrations/science-fair-project-proposal-template-germination-percentage.webp) With 40 seeds per treatment, each germinated seed moves the group's germination percentage by 2.5 percentage points. That is fine granularity for a school project. Write your analysis plan before collecting data so you cannot accidentally choose the analysis that flatters your hypothesis. The core calculation is a percentage and a difference between groups. Here is a small Python script that computes germination percentages from raw counts, with the expected output shown underneath. ```python # germination_analysis.py groups = { "0% salt": {"seeds": 40, "germinated": 36}, "0.5% salt": {"seeds": 40, "germinated": 30}, "1.0% salt": {"seeds": 40, "germinated": 22}, "2.0% salt": {"seeds": 40, "germinated": 12}, } for name, g in groups.items(): pct = 100 * g["germinated"] / g["seeds"] print(f"{name}: {g['germinated']}/{g['seeds']} = {pct:.1f}%") control = 100 * groups["0% salt"]["germinated"] / groups["0% salt"]["seeds"] highest = 100 * groups["2.0% salt"]["germinated"] / groups["2.0% salt"]["seeds"] print(f"Difference, control vs 2.0%: {control - highest:.1f} percentage points") ``` Expected output: ``` 0% salt: 36/40 = 90.0% 0.5% salt: 30/40 = 75.0% 1.0% salt: 22/40 = 55.0% 2.0% salt: 12/40 = 30.0% Difference, control vs 2.0%: 60.0 percentage points ``` Those numbers are illustrative placeholders. You will replace them with your own counts. A spreadsheet works just as well. Use `=B2/A2*100` for the percentage and be careful about one common spreadsheet error. If your denominator cell is blank or zero, Excel shows `#DIV/0!` and Google Sheets shows `Function DIVIDE parameter 2 cannot be zero.` Both mean the same thing. Check that your seed count cell is filled in before the formula that divides by it. ### Step 6. Write the data management plan A data management plan is a short paragraph describing what data you will collect, where it will live, how it will be labeled, and whether anyone else can see it. This is not just an adult formality. Funding agencies increasingly require researchers to include data sharing plans in new project proposals unless there is a strong, clearly justified reason not to [4]. The National Institutes of Health supports researchers with structured planning templates, such as the DMPTool template developed for its HEAL Initiative, which guides users through storing data, standardizing metadata, protecting sensitive information, and choosing appropriate repositories [5]. Metadata means the descriptive information about your data, such as when a measurement was taken, with what instrument, and under what conditions. For a science fair project your plan might read: Raw counts will be recorded by hand in the lab notebook on the day of measurement and transferred the same evening to a spreadsheet named `bean_germination_2026-03-14.xlsx` saved in cloud storage and on a USB drive. Each spreadsheet row will contain the date, pot number, treatment concentration, seeds planted, and seeds germinated. The spreadsheet will be shared with my teacher and can be shared with judges on request. No personal or sensitive information is involved. ### Step 7. Add safety, timeline, and acknowledgments Safety for this project is simple: wash hands after handling soil, keep electrical devices away from watering areas, and do not taste any solution. A timeline might run two weeks from proposal approval to data collection, one week for analysis, and one week for the display board. Acknowledge everyone who helped and say how, for example "My teacher suggested the root-length criterion. My mother drove me to buy seeds." ## Building Your Lab Notebook Alongside the Proposal The proposal and the lab notebook are two halves of one system. The proposal says what you will do. The notebook proves you did it and records what actually happened, including the surprises. Professional research data management has been moving from loose documents toward structured, traceable records. One framework from materials science combines what its authors call the STAR paradigm, emphasizing Statefulness, Traceability, Aim, and Result, with a methodology for Standardisation, Extraction, and Testing of research data, and shows that this progression helps research outputs consolidate into reusable datasets [6]. You do not need that machinery, but the underlying ideas transfer directly to a school lab notebook. Write the aim at the top of each day's entry. Record raw numbers as you observe them, never from memory at the end of the week. Note anything unexpected, such as one pot drying out overnight. Sign and date each page. If you use a lab notebook template PDF, print one page per experimental session with blanks for date, aim, conditions, observations, and a small results table. Metadata deserves special attention. A group of biomedical researchers proposed a minimal metadata set, a short list of information that must accompany raw data so it can be understood and reused later, aligned with the ARRIVE 2.0 reporting guideline for in vivo research [7]. Their point was that data collected without adequate metadata cannot be fully interrogated or integrated into later work, which wastes resources. Translated to your project, that means every row of your spreadsheet should carry its date, its pot identifier, and its treatment, because in March you will not remember which unlabeled column of numbers was which. The same reasoning underlies the FAIR principles, a widely adopted set of standards holding that research data should be Findable, Accessible, Interoperable, and Reusable [8]. Community-driven recommendations built around those principles emphasize stewardship, training, and support as the practical foundations of good data practice [8]. For your purposes, a notebook and spreadsheet that a classmate could understand without asking you a single question meets the spirit of all four words. ## Doing Background Research the Right Way The background section of the proposal should show that you looked for existing answers before planning your experiment. Use at least three sources, prefer ones aimed at scientists or students of science over generic websites, and write two or three sentences on what each source told you. Record the full citation for each. There is one professional habit worth borrowing here. When systematic reviewers, the researchers who produce exhaustive summaries of medical evidence, cannot find data in published papers, they sometimes write directly to the organizations that hold the missing results. The Drug Effectiveness Review Project did this for nine years, receiving 160 dossiers from pharmaceutical manufacturers covering 40 reviews, and found that most submissions were unique studies and nearly 42 percent were unpublished [9]. The yield of actually usable material was small, only 2.2 percent of 7,360 submitted studies or datasets made it into a review [9]. The transferable lesson is that going to the original source beats relying on secondhand summaries, and that asking an expert a specific question often works better than searching harder. If your project involves a local factor, such as soil chemistry or water quality, a short email to a university extension office or a municipal lab can give your background section real authority. ## How Reviewers Actually Judge Proposals Understanding the review process changes how you write. Science fairs, grant agencies, and open research communities all face the same problem. They have more worthy projects than resources, and they must decide quickly and fairly. One open science community tested an alternative called community review. Across 147 projects in two open research initiatives, the process produced 614 reviews from 201 reviewers over 7 funding rounds and awarded 48 micro-grants of up to 4,000 euros. It was fast, with a median process duration of 10 days, and it stayed fair, with project rankings holding up even when individual reviewers were mathematically removed from the analysis [10]. The design feature most relevant to you is granular review forms. Instead of one overall impression, reviewers scored specific dimensions. A proposal that addresses each dimension separately, with its question, method, analysis, and safety clearly separated, gives a reviewer no blank boxes to guess at. Clarity matters even more than most students expect, because expert judgment is noisier than people assume. A group preparing a large clinical and translational science grant application tried to build a simple instrument that reviewers could use to reliably distinguish two related categories of research. Even experienced faculty reviewers reached only fair agreement, with median kappa coefficients of 0.39 and 0.22 across the determinations, and the detailed algorithm did not improve agreement [11]. If trained professionals reviewing published papers disagree at that rate, your safest strategy is to remove every judgment call. State the outcome measure, the number of trials, and the success criterion in numbers, so that no reviewer has to interpret what you meant. Bias is the other documented reality of review. The proposal-free funding model mentioned earlier was motivated partly by the overhead and bias of conventional grant peer review, and its simulations over two decades of citation and funding records suggested it could produce distributions similar to those of the National Science Foundation and the National Institutes of Health while possibly being more equitable [1]. You cannot redesign the system, but you can write in a way that survives it. Concrete numbers, named controls, and an explicit analysis plan give every reviewer, agreeable or skeptical, the same facts to work from. ## Group Projects and Crediting Help If you are working with a partner, decide before you write the proposal who did what, and write it down. Professional research takes this seriously because credit disputes are common and corrosive. An analysis of authorship ethics in global health research partnerships identified four recurring problems, including the fact that standard authorship guidance can systematically exclude contributors whose main work was not writing or revising the manuscript in English, and that power differences between senior and junior partners can distort who appears in which position [12]. The same paper notes that purely technical contributions, such as building the apparatus or writing the code, receive insufficient recognition under conventional rules [12]. Your proposal's acknowledgment section is the place to fix this in advance. Name every helper and describe the contribution, for example "Design of the watering schedule was shared equally" or "Statistical advice came from my uncle, a statistics teacher." ## Common Mistakes and Limitations These are the errors reviewers report most often, with the fix for each. - Topic instead of question. "Studying plants and salt" cannot be approved or rejected. Replace it with a sentence ending in a question mark that names the organism, the variable, and the measurement. - No control group. If every condition is new, you cannot tell whether the change you observe would have happened anyway. Always include one condition that represents normal or baseline conditions. - More than one independent variable. Testing salt concentration and light level at the same time makes the result uninterpretable. Change one thing, hold the rest constant, and list them in the variables section. - A hypothesis that cannot fail. "The project will be interesting" predicts nothing. Write a directional prediction tied to the measurement, such as "germination percentage will decrease as salt concentration increases." - Vague materials. "Some soil, some seeds" fails the repeatability test. Give brand names where relevant, counts, volumes in milliliters, and masses in grams. - Sample size too small. One plant per treatment tells you about one plant, not about seeds in general. Forty seeds per treatment, as in the worked example, gives each seed a 2.5 percentage point influence on the result, which is reasonable for a school project. - No data plan at proposal time. Deciding what to record after the experiment starts guarantees gaps. Attach your notebook page format or spreadsheet column headers to the proposal itself. - Analysis chosen after seeing results. Decide in the proposal that you will compare percentages across groups. Choosing the statistic that flatters your data afterward is a form of bias, and judges recognize it. - Copying the background from one website. Reviewers recognize pasted text instantly. Three short paraphrased summaries with citations beat three pages of copied paragraphs. - Ignoring the timeline. A proposal with no dates cannot be checked for feasibility. Put real calendar dates next to each phase. One honest limitation belongs in your proposal too. A single school experiment with 160 seeds in one classroom supports a limited conclusion about those seeds under those conditions. Real research validity depends on repeating measurements across conditions and on records detailed enough for others to verify, which is exactly why the notebook and data plan sections exist in the template. State what your experiment can and cannot show, and reviewers will treat the claim seriously. Individual animal, human, or clinical questions raise additional rules that vary by fair and by institution, and your adult sponsor or fair's safety committee is the right authority for those cases. ## Frequently Asked Questions ### What is a science fair project proposal template? It is a preformatted document with labeled sections for the question, hypothesis, variables, materials, procedure, data plan, safety, and timeline of a planned project. You fill in each section rather than inventing a structure, which keeps your proposal complete and comparable to what reviewers expect. ### How long should a science fair proposal be? Most school-level proposals run two to four pages. Follow your fair's stated limit if it has one, and prioritize completeness of the required sections over total length. ### Do I need a hypothesis before writing the proposal? Yes. The hypothesis drives the variables, the sample size, and the analysis plan, so reviewers expect it in the proposal. If your question is genuinely exploratory, state that explicitly and describe what observations you will collect. ### What is the difference between the proposal and the final report? The proposal is written before the experiment and predicts what will happen. The final report is written after and presents what actually happened, including results that contradicted your hypothesis. A well-filled proposal becomes roughly half of the final report with minimal rewriting. ### Where can I find a free science fair project proposal template? The template in this article is free to copy and adapt, and most regional and school fairs publish their own required forms on their official websites. Use your fair's form when one exists, and use the structure here as your working draft. ### Do I need a lab notebook, and what is a lab notebook template? Yes, a lab notebook is where raw observations go, in ink, dated, at the time of observation. A lab notebook template PDF is a preprinted page with blanks for date, aim, conditions, and results that keeps entries consistent. A bound notebook is preferred over loose pages because page order cannot be rearranged. ### What belongs in a data management plan for a school project? Four things: what data you will collect, where the files will be stored and backed up, how files and records will be named and labeled, and whether the data will be shared with anyone. Professional requirements increasingly mandate such plans in research proposals, and the habit costs you one paragraph. ### Can I change my project after the proposal is approved? ![Decision path asking whether a change touches the question, procedure or safety, leading to a note to the sponsor or an updated proposal.](/article-images/illustrations/science-fair-project-proposal-template-change-after-approval.webp) Minor changes in materials or timing usually require only a note to your adult sponsor. Changes to the question, the procedure, or anything with safety implications require an updated proposal and re-approval. Ask your sponsor the same day the need for a change becomes clear rather than after the fair deadline. ## Sources 1. [An efficient system to fund science: from proposal review to peer-to-peer distributions.](https://pubmed.ncbi.nlm.nih.gov/29795961/) 2. [How reliable is peer review? An examination of operating grant proposals simultaneously submitted to two similar peer review systems.](https://pubmed.ncbi.nlm.nih.gov/9393374/) 3. [Refining Established Practices for Research Question Definition to Foster Interdisciplinary Research Skills in a Digital Age: Consensus Study With Nominal Group Technique.](https://pubmed.ncbi.nlm.nih.gov/39847774/) 4. [Context of data sharing practices in collaborative human genomic research in low and middle income countries: A systematic review.](https://pubmed.ncbi.nlm.nih.gov/42566494/) 5. [Navigating the new NIH Data Management Sharing plan: the NIH HEAL Initiative Data2Action DMPTool template.](https://pubmed.ncbi.nlm.nih.gov/40401004/) 6. [Evolve with your research: stepwise system evolution from document-driven to fact-centric research data management in materials science.](https://pubmed.ncbi.nlm.nih.gov/41845429/) 7. [A minimal metadata set (MNMS) to repurpose nonclinical in vivo data for biomedical research.](https://pubmed.ncbi.nlm.nih.gov/38438748/) 8. [Recommendations for Services in a FAIR Data Ecosystem.](https://pubmed.ncbi.nlm.nih.gov/33205119/) 9. [A study of the value of requesting information from drug manufacturers for systematic reviews; 9 years of experience from the drug effectiveness review project.](https://pubmed.ncbi.nlm.nih.gov/30348228/) 10. [Community review: a robust and scalable selection system for resource allocation within open science and innovation communities.](https://pubmed.ncbi.nlm.nih.gov/38283124/) 11. [Steps toward developing an algorithm to facilitate recognition of translational science.](https://pubmed.ncbi.nlm.nih.gov/42375221/) 12. [Authorship ethics in global health research partnerships between researchers from low or middle income countries and high income countries.](https://pubmed.ncbi.nlm.nih.gov/24885855/)