| New Hampshire Science and Engineering Exposition (NHSEE) |
How to Build a Project Abstract |
🔬
The Scientific MethodFor hypothesis-driven experiments Step 1 — Introduction
Why does this matter?
Identify the topic you are investigating and explain why it is important. Who is affected? Why should someone care about your findings?
Step 2 — Question & Hypothesis
What do you predict?
State your specific research question. Then form a hypothesis — a testable prediction in "if...then...because" format — based on your background research.
Step 3 — Methods
How did you test it?
Design a controlled experiment. Identify your independent variable (what you change), dependent variable (what you measure), and control group. Repeat trials to ensure reliable results.
Step 4 — Data & Analysis
What did the data show?
Record and organize your measurements. Look for patterns and trends. Use graphs, tables, and statistical analysis to make sense of your results.
Step 5 — Conclusions
Was your hypothesis supported?
Explain whether your results supported or contradicted your hypothesis. Acknowledge sources of error and what you would do differently in a future experiment.
Step 6 — Applications & Implications
Why does your work matter?
Connect your findings to the real world. Who could benefit from what you discovered? What future research directions does your project suggest?
| ⚙️
The Engineering Design ProcessFor design, build, and test projects Step 1 — Introduction
What problem are you solving?
Identify the problem or need you are addressing. Who is affected by this problem? Why is it worth solving? Clearly define the scope of what you set out to do.
Step 2 — Design Goal & Initial Design
What did you set out to build?
State your design goal and the criteria your solution must meet (what it must do) and constraints you must work within (budget, size, materials). Describe your first design concept.
Step 3 — Build & Test
How did your first prototype perform?
Build your initial prototype and test it against your design criteria. Collect data on its performance. What worked? What failed? Document your results carefully.
Step 4 — Redesign & Iterate
How did you improve your design?
Use your test results to improve your design. Rebuild and retest — this cycle of designing, testing, and refining is the heart of engineering. Track what changed and why across each version.
Step 5 — Final Design & Results
What does your final solution look like?
Describe your finished design and how it performs against your original criteria. Include specific test data from your final prototype. How does it compare to your very first version?
Step 6 — Applications & Implications
What is the real-world impact?
Who could benefit from your solution? Could it be manufactured, scaled, or deployed more broadly? What would you improve in a future version, and what new problems does your work raise?
| 💻
The Computational MethodFor algorithms, models, and data analysis Step 1 — Introduction
What problem are you tackling?
Identify the real-world problem or question you are addressing with computing. Why is it important? What gap in knowledge or capability does your project aim to fill?
Step 2 — Problem Statement
What exactly are you trying to do?
Define your specific research question, hypothesis, or computational goal. Be precise — a strong problem statement explains exactly what you are building, predicting, or analyzing and what success looks like.
Step 3 — Computational Methods
How did you build or analyze it?
Describe the tools, languages, libraries, algorithms, and datasets you used. Include your model architecture or analysis pipeline, key parameters, and how you validated your approach (e.g., train/test split, cross-validation).
Step 4 — Key Results
What did your system or analysis produce?
Report your most important quantitative findings — accuracy, precision, recall, error rates, simulation outputs, or other metrics. Compare your results to a baseline or prior work when possible.
Step 5 — Conclusions & Implications
What do your results mean?
Interpret your findings. Did you meet your goal? What are the real-world implications of your results? Identify the limitations of your approach and describe what future work could extend or improve your project.
|
Abstract Builder |
Science Fair Abstract BuilderAnswer each question to build a complete, well-structured abstract for your project. |
Tips for Building an Abstract |
Writing an abstract The Abstract Builder above can help you organize your ideas by breaking each part of the abstract down my section. If you are doing a wet-lab experiment where you are collecting data, select the science builder. If you are designing an application or a prototype, select the engineering builder. If you are doing data analysis or modeling from existing datasets, use the computational builder. You have 250 words (ISEF rule) to draw your reader into your paper and convince them to read the whole thing. A Quick Guide to Abstracts What is an abstract? An abstract is a brief, written description of your project that explains your project’s purpose, procedures, data, and conclusions. It is a self-contained summary that tells the reader why they should care about your project and what you found out. The abstract is concise, but complete—it communicates the essence of your project. How do I write an abstract? ISEF limits abstracts to 250 words, so you need to be succinct—focus on the big picture. Here are a couple of templates: Science Experiment. For a science experiment, start with an introductory statement about why you are doing your project. What are you trying to find out, and why should your readers care? Then, state your question/problem and your hypothesis. Next, summarize your data collection methods. You don't have a lot of space, so only mention the key points of your procedures. Then give the highlights of your data and data analysis, followed by your conclusions. The very last part of the abstract should discuss the applications and implications of your project. Engineering Project. The abstract for an engineering project will be similar, but there will be a few changes. Start with an introductory statement about why you are doing your project. What problem are you trying to solve? What need are you trying to address? Why should your readers care? Then, state your design goal and describe your initial design. Next, summarize your iterative process of designing, testing, rebuild, and retesting. You don't have a lot of space, so only mention the key points of your process. Then give the highlights of your data analysis from tests of prototypes, followed by your final designs. The very last part of the abstract should discuss the applications and implications of your project. Computational Project. For a computational project, start with an introductory statement about why you are doing your project. What are you trying to find out, and why should your readers care? Then, state your question/problem and your hypothesis. Next, summarize your computational methods. You don't have a lot of space, so only mention the key points of your procedures. Then give the highlights of your data and data analysis, followed by your conclusions. The very last part of the abstract should discuss the applications and implications of your project. Are Abstracts Important? Yes, very! Abstracts are an incredibly important part of technical communication. Scientists, engineers, and mathematicians have to wade through a vast amount of literature. They don't have time to read all of it, so they use abstracts to decide if an artic le is worthwhile. If the abstract is interesting and relevant, a scientist might read the accompanying article. If the abstract isn’t, a scientist stops reading and moves on. Just like scientists, science fair judges read your abstract and may make preliminary judging decisions based on your abstract. A good abstract is like a good first impression it goes a long way. For the science fair, the abstract is worth 5 points of your total score. Do you have any other tips?
|