Parent Guide · KSEF — Korea Science & Engineering Fair

Before running the experiment,
we set the research question

KSEF (Korea Science & Engineering Fair) is an inquiry competition where the student sets a research question and creates new data or experiments to answer it, then explains it in a report and a presentation. It is organized by Korea Science Service. Good research begins with a good question, and sharpening that question is itself a process. This page shows that research-ideation process, in pictures.

Overview Summer-centered · report (~10 pp) + presentation · 1:1 or team Categories, eligibility, and dates: check the official notice each year
01 · What's judged

The report and the presentation, together

KSEF looks not at a single result but at the quality of the research and how well the student understands it. So as much as what was found, it matters which question was set and how it was answered.

The report and presentation are judged together. The student must do the work themselves and be able to explain every method and conclusion independently.

Inquiry contest
The student sets a question and creates new evidence
Report + presentation
Deliverables — judged together
Annual notice
Categories, eligibility, dates: per official notice

On international pathways, honestly: nomination to international events such as ISEF is decided by each year's official notice and is not automatic. Awards aren't guaranteed either. Still, the experience of setting your own question and seeing a study through becomes a foundation for what comes next.

02 · Research is a process too

Widen, then narrow — twice

A good research question does not arrive all at once. It goes through four set steps. First an interest is explored wide and narrowed to one question; then methods are opened wide and narrowed to one workable design.

1 2 3 4 EXPLORE QUESTION DESIGN COMMIT
◇ Question spaceMethod space ◇
Step 1 · Explore

Explore interests

Look broadly at phenomena and topics you're curious about; collect questions without judging.

Widen
Step 2 · Question

Research question

Sharpen one question worth pursuing into something measurable and narrow.

Narrow
Step 3 · Design

Design methods

Sketch several ways — experiment or data collection — to answer that question.

Widen
Step 4 · Commit

Commit the design

Settle on one method you can actually carry out over the summer.

Narrow

If the experiment hasn't started in the first weeks, that's normal. This is the stage for finding and narrowing the question. A study begun with a fuzzy question stalls at the report.

03 · The four steps of KSEF

From choosing a topic to the presentation

Once the question is set, the real research follows. The four steps below are the backbone of KSEF prep. The field is chosen not for "how it looks" but for an interest the student will pursue to the end — research is long, and without interest it stalls at the report.

Categories, eligibility, and document formats change every year. Once the interest is set, check that year's requirements on the competition page and the official notice (ksef.or.kr).

04 · Making the evidence yourself

Not found material — made evidence

The core of KSEF is that the student creates new evidence. Rather than summarizing existing material, they read prior work to find a distinct angle, make data by measuring or experimenting, and analyze it. Four tasks follow.

📚

Read prior work

Find similar studies, note what's already known, and pin down what makes your question different — a narrow gap no one has answered yet.

Papers · prior work
🎛️

Design variables · measurement

Decide what you change (independent), what you measure (dependent), and what you hold equal (controlled). The fairness of the experiment is set here.

Experiment design
🔬

Collect data · experiment

Make new data by measuring, experimenting, or surveying. For coding/data topics, the student writes the code and can explain the process.

Hands-on
📈

Analyze

Interpret results with tables, graphs, and statistics, and see whether they answer the question. Record unexpected results as they are — new findings come from there.

Statistics · analysis

Rule for this stage: don't decide the conclusion in advance. Don't bend the data to the result you want; write what the data says. Doing the work yourself and recording every step reproducibly is the heart of KSEF and ISEF.

05 · What a good research question looks like

Same interest — the depth of the question is what matters

Narrowing a vague interest into a measurable, reproducible question is half the research. With the same interest, how well you've honed the question decides whether it becomes research or an unworkable claim.

Vague — not research"AI is useful." A claim, not a question. Nothing to measure, no variable to change.
Better — still broad"I want to improve image-classification accuracy." A direction, but no what, and under which conditions, is being measured.
Testable"How does the accuracy of a model estimating food-waste amount from cafeteria-tray photos change with lighting (illuminance)?" The measurement target, the variable changed, and reproducible conditions are all clear.

Writing the hypothesis

If [what you change · independent variable] changes, then [what you measure · dependent variable] will ~ — because [reason].

e.g. If the lighting is lowered, food-waste estimation accuracy will drop — because in darker photos food edges blur and the model distinguishes waste less well.

Tell the three variables apart

Independent

What you change. e.g. lighting.

Dependent

What you measure. e.g. estimation accuracy.

Controlled

What you keep equal for fairness. e.g. camera, distance, food type.

06 · From question to conclusion

Question → method → analysis → conclusion

A well-set question is the map for the whole study. The method follows from the question, the analysis from the method, and the conclusion from the data. Read left to right.

Research question → method · data → analysis → conclusion · limits
QuestionHow does food-waste estimation accuracy change with lighting?
Method·dataPhotograph and measure trays at set lighting levels; collect model estimates
AnalysisCompare error across lighting bands with statistics and graphs
Conclusion·limitsAccuracy drops sharply below a certain illuminance; state the need for correction and its limits

The shape of a study that goes the distance

Score each candidate question on five axes; if any one is weak, narrow or change the question. A strong study fills all five evenly.

Question Method Repeatable Data Novelty
A study that goes the distance A weak question

The five axes are question clarity · method validity · data grounding · originality · reproducibility. However interesting the topic, if one axis collapses it's hard to finish over the summer.

Re-narrow criteria (e.g.): if the data can't be collected within the summer, the variables can't be controlled, or the conclusion is already widely known, re-hone the question. Narrowing this way settles on one question you can actually do.
07 · So here's the schedule

The front end is "honing the question"

Prep centers on the summer break and counts back from that year's official deadline. The red and gold bands below are finding and narrowing the question, the green band is making the evidence, and the last is writing up and presenting.

W12345678910111213141516
Topic · pre-research
prior work
Question · design
question → method
Research · experiment
making data
Analysis · report
student-written
Presentation
report + talk
Finding the question Narrowing the question Making the evidence Report · presentation

Fit school exams and assignment deadlines first, then set the research schedule. The first weeks show little, but the question and design honed here carry the whole study. Timing and length vary by topic and by that year's operation.

08 · What your child gains

What stays, apart from an award

This process is not a competition-only skill. Setting your own question, making your own evidence, and seeing it through to explain it carry straight over to coursework and later research and presentations.

Framing a question

Learning to narrow a vague interest into a measurable question.

Making evidence firsthand

Building the habit of creating data yourself and handling it fairly, not borrowing others'.

Seeing it through

Building the stamina to plan a long study and finish it as a report.

Explaining and answering

Presenting the research clearly and answering questions on it.

There is little you need to do at home. Listen to what your child stays curious about, and help arrange the schedule so they can set aside research time over the summer. One of your child's interests becomes the starting point of a research question.

Awards are not guaranteed. Categories, eligibility, documents, dates, and nomination to international events such as ISEF differ each year and are not automatic, so always check that year's official notice (ksef.or.kr). Whether and how it can be submitted to a university depends on each program's rules and must be confirmed separately. For details see the competition page; for questions, contact your teacher or jc@citcoding.com. Same approach for other competitions: CAC · Technovation · KOAI.

CIT Code Academy · Parent Guide — The research-ideation process KSEF · Korea Science & Engineering Fair