Designing Science Inquiry (STEAM) Activities with AI: From Question to Record Sheet
Aligned with the new primary Science curriculum, this article shows how teachers turn a topic into a complete inquiry activity in 20 minutes with AI — question, hypothesis, variables, procedure, safety notes, record sheet and extension — with three worked examples and an inquiry quality checklist.
Short answer: AI can expand one topic into a complete inquiry activity in 20 minutes — question, hypothesis, variables, procedure, safety, record sheet, extension. But the soul of inquiry, letting students decide, must be guarded by the teacher.
What the new curriculum asks: from knowing to inquiring
The primary Science curriculum emphasises inquiry skills: questioning, predicting, designing, observing, recording, explaining, communicating. That is a new challenge for teachers — not just teaching content but designing activities where students "do science". AI cuts design time dramatically so teachers can focus on facilitation.
The seven parts AI generates
| Part | AI output | Teacher check |
|---|---|---|
| 1. Inquiry question | 2–3 investigable, measurable questions | Pick one close to students' experience |
| 2. Hypothesis | Sentence frames for students | Leave blanks for students |
| 3. Variables | Independent, dependent, controlled | Check they are really controllable |
| 4. Procedure | Steps with timings | Trial it yourself |
| 5. Safety notes | Risks and alternatives | Final judgement |
| 6. Record sheet | Tables, chart frames, observation column | Adjust fields |
| 7. Extension | Follow-up questions, cross-subject links, home inquiry | Optional |
Example 1 (P3): Which paper towel absorbs fastest?
- Question: Do different brands absorb water at different speeds?
- Variables: independent = brand; dependent = water height after 30 s; controlled = water volume, towel size, time.
- Procedure: cut to equal size → dip 1 cm simultaneously → measure after 30 s.
- Safety: none specific; keep water away from devices.
- Record sheet: brand × three trials × mean.
- Extension: Why the difference? Related to towel structure?
Example 2 (P5): What affects shadow length?
- Question: How does light-source height affect shadow length?
- Variables: independent = torch height above desk; dependent = shadow length; controlled = object height, distance.
- Safety: torches only; do not look into bright light.
- Record sheet: heights 10/20/30 cm × shadow length; plot a line graph.
- Extension: How do shadows change through the day? Seasonally?
Example 3 (P6): Phototropism
- Question: Do seedlings grow towards light?
- Variables: independent = light direction; dependent = stem tilt angle over 7 days; controlled = water, soil, species.
- Safety: none; observe at the same time daily.
- Record sheet: date × angle × photo.
- Extension: What if the pot is rotated daily?
Inquiry quality checklist (3 minutes)
- Is the question measurable? (Not "why is the sky blue?")
- Is there something students must decide themselves? (hypothesis, controls, how to record)
- Are controlled variables really controllable?
- Could the result contradict the hypothesis? (inquiry must allow failure)
- Are safety notes sufficient? Has the teacher trialled it?
Save to the panel library
Good inquiries belong in the panel resource library with classroom photos and sample student records, tagged by grade and topic. Next year's teacher takes it in one click and tweaks it for their class — that is how panel knowledge accumulates.
Takeaway
AI makes inquiry design no longer the panel head's secret craft. Enter a topic, get a full activity and record sheet in 20 minutes — then spend the time on what matters most in class: listening to students' hypotheses and asking "why do you think so?"
FAQ
The AI lists safety notes and alternatives, but the final safety judgement is the teacher's. Any activity involving heat, electricity, chemicals or sharp objects should be trialled by the teacher first.
In inquiry, students pose or choose the question, hypothesise, decide variables and explain results; a follow-the-steps experiment only verifies. Good AI output keeps the parts students must decide, rather than fixing everything.
Yes. Ask the AI to generate under a "classroom-feasible" or "home-feasible" constraint using everyday materials; many primary inquiries can be done on a classroom desk.
Related articles
AI Lesson Planning in Practice: A Lesson Plan and Tiered Worksheets in 30 Minutes
Using P4 Chinese (expository writing) and S2 Mathematics (linear equations) as examples, see how teachers start from a school curriculum unit and use AI to produce a lesson plan, slide outline, three-tier worksheets and reflection prompts in 30 minutes — with a reusable prompt structure.
Read moreThe Scaffolded AI Tutor: Why an AI That Won't Give Answers Is the Right One for Students
The 4-level hint ladder, Socratic questioning and teacher-set "hint ceiling" behind a scaffolded AI tutor — why answer-giving AI weakens learning, with dialogue examples from primary maths and secondary English.
Read moreAI Question Generation, Marking and Instant Feedback: Practice and Limits in Maths, Chinese and English Writing
How AI marking works in three subjects — step-by-step maths marking (including handwritten photos), Chinese composition feedback against a rubric, English writing grammar and structure suggestions — plus the three checkpoints teachers must keep, with a rubric template.
Read more