Will my child be tested, or interviewed, or both?

It depends on the school and often on the specific talent area within it. Published material describes a two-hour written problem-solving paper at one school, selection tests then a camp at another, a selection test then an interview at a third, and one school whose table says shortlisted mathematics, science and robotics applicants are contacted for interviews or trials — the school choosing which. Read each school's own page.

Does my child need competition awards to apply?

Not universally, but some talent areas publish hard gates. One school states that its mathematics talent area considers only Platinum awardees of a particular olympiad round. Other schools publish that applicants without prior experience may apply and will be assessed on potential. These are opposite policies at different schools, which is why the school's own document is the only source worth acting on.

How do panels tell whether a child really did the project?

By questioning specifics — why a design decision was made, what a notebook entry meant, what was tried before this. That is a synthesis published by a preparation provider rather than a first-hand account, so treat it as a well-argued hypothesis. It is also a good description of how anyone would check, and preparing for it costs nothing.

What if the project was a team project?

Say so, clearly, and name your own part without shrinking it or inflating it. Robotics in particular is commonly team-based, and panels know this. The failure mode is a child who uses “we” for everything and cannot say what they personally did.

Are there group tasks in STEM selection?

Yes at some schools, and they are published. One school runs a group STEAM challenge as its second phase; another runs a multi-day selection camp for shortlisted candidates. What happens inside those rooms is almost entirely undocumented — no first-hand account of one exists in the public record, and any provider claiming to simulate the real thing is over-claiming.

Should my child memorise a script for the project talk?

No. Two independent sources with opposing commercial interests — a coach who sells preparation and a parent who had been through selection — both say memorised answers come apart as soon as the question shifts. Practise the material until it is familiar, not until it is fixed.

Is a STEM DSA-Sec place a commitment to a STEM CCA?

Published school documents state that a student admitted through DSA-Sec must take part in activities related to the talent they were selected for from Year 1 to Year 4, cannot take part in S1 Posting, and cannot transfer after PSLE results. For a child whose interest is genuinely wide, four years pointed at one area is worth thinking about before applying.

Science, maths and engineering DSA-Sec Interview Questions and Preparation

What is being assessed

What does a science, maths and engineering DSA-Sec panel actually look for?

A science, maths and engineering panel is checking authorship and reasoning. It wants to hear the method, the alternatives that were rejected, and at least one thing that went wrong — because failure is the part of a project that cannot be borrowed from anyone else.

Whether the child, not a coach or a parent, made the design decisions. Questioning about why a specific choice was made is the reported mechanism for this, and it is why a polished summary is less convincing than a rough explanation with real detail in it.
Whether they can explain a method to someone who does not already know it. Panels are not always specialists in the child's exact sub-field.
Whether they can describe a dead end. A project with no failed attempt in it is either unusually lucky or incompletely remembered.
How they think aloud on an unfamiliar problem. Live reasoning tasks — a novel problem, explain your approach — are reported at more than one selective school and published in a group form by at least one.
Genuine curiosity outside the syllabus. At least one school publishes that its selection test includes general-knowledge questions intended to gauge how widely read the student is in mathematics and the sciences.

The project matters less than whether your child can explain it — and say what went wrong.

The field

Which secondary schools offer science, maths and engineering as a DSA-Sec talent area?

Science, technology, engineering and mathematics is offered by 68 of the 143 secondary schools publishing DSA-Sec talent areas — 106 talent-area entries across 23 distinct labels. It is a mid-sized field with unusually fragmented naming, and that fragmentation matters far more than the headline number when you are deciding where to apply.

Schools offering STEM

68 of 143

PSLEPrep extraction from MOE SchoolFinder, 8 August 2026

Distinct STEM labels

From robotics and mathematics down to single-school areas

IP schools with STEM

Integrated Programme schools offering at least one STEM talent area

Schools admitting only via DSA

2 STEM-focused

NUS High and the School of Science and Technology take 100% of intake through DSA-Sec

The fragmentation matters when you build your school list. Robotics and a generic “STEM” label appear 21 times each, mathematics and science ten times each, and then it splinters: environmental science, coding, digital media, computational thinking, programming and 3D modelling, aerospace, forensic science, mechatronics. Two schools admit their entire Secondary 1 intake through DSA-Sec on a STEM basis, so for them this is not one route among several — it is the only door.

Reported questions

What questions come up in science, maths and engineering DSA-Sec interviews?

Reported STEM questions split into two families: broad curiosity questions about what interests the child in the subject, and narrow authorship questions about specific decisions in their own submitted work. The second family is the one that decides the session.

How to read the labels

Confirmed: A school has published this task or probe in its own DSA document. Only that it exists — and what the school says it is for — is confirmed. The wording below is ours.
Corroborated: Described first-hand by a parent or candidate who was in the room. Anecdote, but from someone who was actually there.
Unverified: Circulates in tuition question banks or guides with nothing behind it, or rests on a single second-hand report. Worth rehearsing. Not evidence that any school asks it.

The bank is short for this area, and that is the finding rather than a gap in our research: families who go through selective STEM processes post about them least, so the public record is thinner here than for sports or the arts. The three below are the ones tagged to science, maths and engineering, and the third is the live reasoning task.

“Which part of this piece was challenging for you?”
Corroborated
Why they ask it:
Two things at once — self-critique, and authorship. Somebody who made the work knows exactly which corner fought back.
The weak answer:
Saying none of it was hard. It reads as either untrue or incurious, and it is the fastest way to make a panel wonder how much of the piece was the child's.
Where this comes from:
Reported first-hand in the parent account of visual-arts sessions, and structurally similar to the portfolio questioning described for technology talent areas.
“Why did you make that choice in your project? Walk us through how you decided.”
Unverified
Why they ask it:
Authentication. If a coach or a parent did the heavy lifting on a submitted project, the design decisions are where it shows.
The weak answer:
Describing what the project does instead of why it was built that way. The panel already saw what it does. Naming an option that was rejected, and why, is what proves the work.
Where this comes from:
Described in a prep-service guide as a synthesis of school criteria and community reports — not a verbatim question, and not first-hand. Treat the pattern as plausible and the wording as ours.
“Here's a problem you haven't seen before. Talk us through how you'd approach it.”
Unverified
Why they ask it:
Reasoning out loud, which is a different skill from getting the answer. Live tasks like this cannot be pre-drilled, which appears to be the point.
The weak answer:
Going quiet to think. Silence is the one thing the assessor cannot mark. Saying what you notice first, and where you might be wrong, is the whole task.
Where this comes from:
Reported by a prep-service guide as the shape of live assessment at some academically selective schools. It is the author's synthesis, not a first-hand account of any session.

A few more, specific to this talent area

A few more that circulate specifically in this area. Almost all of it is preparation-industry material rather than anything anyone recorded in a room, and the tiers say so.

“Which mathematical idea interests you the most, and why?”
Unverified
What it is probing:
Whether interest extends past what was taught. A short, specific answer beats an ambitious one.
Where this comes from:
Commercial question bank presented as drawn from case studies, with nothing attributed to a school, year or candidate.
“What would you change about the science curriculum?”
Unverified
What it is probing:
Whether the child has an opinion of their own about how they learn.
Where this comes from:
Same commercial source. No first-hand account was found.
“Talk us through this entry in your notebook — what were you trying at that point?”
Unverified
What it is probing:
Whether the working record is genuinely the child's, and whether they remember the middle of the project as well as the end.
Where this comes from:
A synthesis published by a preparation provider drawing on school criteria and community reports. Reported as a pattern rather than as a question anyone recorded, and the provider sells preparation.
“What did you contribute to this project, specifically?”
Unverified
What it is probing:
Team projects are common in robotics. Being able to name your own part without diminishing anyone else's is the skill.
Where this comes from:
Editorial question list from a parenting publication, repeated across several sites. No first-hand origin found.
“What have you taught yourself outside school — a skill or a topic nobody set you?”
Unverified
What it is probing:
Self-direction, which is the trait most STEM programmes say they select for.
Where this comes from:
Commercial question bank marketed as real questions, with no attribution whatsoever.

Questions about the child themselves, and the “why our school” and rank-order questions, apply to every talent area. They live on the main question bank rather than here.

The process

What are the selection stages for science, maths and engineering DSA-Sec applicants?

STEM is the talent area most likely to begin with a written test rather than a conversation. Published school material describes selection tests, timed problem-solving papers, group challenges and multi-day camps — often with the interview last, and sometimes with no interview at all. At two schools this route is the only way in.

1.A written or selection test
Confirmed
NUS High publishes that all applicants sit selection tests in mathematics and science, assessing understanding and application of what has been taught in schools plus general-knowledge questions to gauge how widely read the student is; only shortlisted candidates go further. Raffles Institution publishes that its STEM domain is assessed by selection test followed by interview. The School of Science and Technology publishes a first phase written test of roughly two hours on applying STEM knowledge to real-world problems.
Read from each school's own published admissions material. Re-check the current cycle's page before relying on any date or format.
2.A project or portfolio talk-through
Unverified
Portfolio submission followed by trials or interviews is reported at several schools, with detailed questioning on design decisions described as the mechanism for verifying the student did the work. This is a synthesis by a commercial provider rather than a first-hand account, so treat the shape as plausible and the detail as unconfirmed.
Commercial synthesis plus scattered forum reports. Weakest evidence on this page.
3.A group task or challenge
Confirmed
The School of Science and Technology publishes a second phase “STEAM Challenge” in which candidates work in groups on real-world problems. NUS High publishes a two-phase process ending in a selection camp for shortlisted candidates only. A parent reported one school running a group task and an interview on the same day for its STEM applicants.
School-published material for the first two; a single forum account for the third.
4.The interview
Confirmed
Where an interview exists it usually sits at the end. One school publishes a talent-by-talent table showing that mathematics, science and robotics applicants are contacted for interviews or selection trials, with the school choosing which — so an interview is not guaranteed even in this area. Hard entry gates are also published: one school considers only Platinum awardees of a particular mathematics olympiad round for its mathematics talent area.
Read directly from a school's published DSA-Sec document for the 2025 exercise. Gates change between cycles.

Formats vary far more than most families expect. The DSA-Sec interview guide covers what happens in the room across every talent area.

Worked answers

How should your child actually answer these?

Three things carry a STEM session: a ninety-second explanation of the project that a non-specialist can follow, a specific decision your child can defend, and one honest account of something that did not work. None of the three can be borrowed from the adult who helped, which is exactly the point of them.

The project in ninety seconds

Tell us about your project.

Problem, what you built, what it does now, in that order, with no jargon that has not been explained. The test is whether an intelligent adult who does not know the field can follow it. If your child cannot explain it to you, they cannot explain it to a panel.

“My grandmother keeps forgetting whether she's taken her tablets, so I made a pill box that lights up. There's a sensor in each compartment that notices when the lid opens, and a small board keeps the time. If a compartment hasn't been opened by eight in the evening, that light turns red. It's not connected to anything — I tried to add a phone alert and I couldn't get it working.”

The last sentence is doing most of the work. A child who volunteers the limit of their own build sounds like the person who built it.

The decision you can defend

Why did you do it that way?

Pick one real choice — a sensor, an algorithm, a material, a variable you controlled — and be ready with the alternative you rejected and why. One defended decision is worth more than a complete tour.

“I used a reed switch instead of a light sensor because the box sits in a drawer, so a light sensor would think it was closed even when it was open. The reed switch only cares about the magnet. It's less accurate about how long the lid was open, but I didn't need that.”

Notice the trade-off at the end. Naming what a choice costs is the clearest signal of genuine understanding a child can give.

Thinking aloud on something unfamiliar

Here is a problem you have not seen. How would you approach it?

Say the first thing you notice. Say what you would try. Say what would tell you it was wrong. Silence reads as being stuck; narration reads as thinking. Arriving at the right answer is not the assessed part.

“I'm not sure yet. The first thing I notice is that the numbers double each step, so I'd guess it's something like powers of two. I'd test it with the small cases first — if it works for three and four, I'd try to see why. If it broke at five, I'd know the doubling isn't the real rule.”

Practise this with problems your child cannot solve. The point is to make being stuck sound like work rather than failure.

The common failure

What is the weakest common answer in science, maths and engineering DSA-Sec interviews?

The weakest common answer in this area is the finished-product tour: a smooth, complete description of a working project with no decisions in it, no alternatives considered, and nothing that went wrong. It is the answer most likely to trigger the questioning it is meant to avoid.

The finished-product tour

“Our project is an automated hydroponics system. It monitors pH and nutrient levels and adjusts them automatically. We won a merit award at the national competition.”

Why it falls flat. It is a description, not an account. Nothing in it could not have been written by the adult who supervised. Because verification questioning is reportedly the standard mechanism in this talent area, a polished summary tends to invite exactly the sceptical follow-up it was meant to prevent — and a child who has only rehearsed the summary has nothing behind it.

What to do instead. Lead with the problem and one decision. Volunteer one failure before being asked. Say what the build still cannot do. Children resist this because it feels like admitting weakness; in this area it is the strongest available move.

This week

What should your child practise this week?

Practise explaining, not presenting. Have your child teach the project to someone outside the field, defend one decision, narrate a problem they cannot solve, and re-read their own notebook so the middle of the project is as fresh as the end.

None of this needs equipment or a tutor. All of it needs a listener who will ask why.

1.Teach it to a non-expert
Ask your child to explain the project to a grandparent or a younger sibling until they can follow it. Every piece of jargon that survives has to be explained in the same breath.
2.Defend one decision
Pick a single choice in the build and ask why three times in a row. The third why is where genuine understanding either appears or does not.
3.Narrate an unsolved problem
Give them a puzzle above their level and ask them to talk while they work, not after. Ten minutes, twice in a week. The goal is to make thinking audible, not to solve it.
4.Re-read the notebook
Projects are remembered from the end backwards. Reading the working record restores the middle — which is where the interesting questions live.
5.Find out what each school actually assesses
Several schools publish their STEM selection format precisely: selection test, group challenge, camp, interview or a combination. Two admit their whole intake this way. Read those pages before you decide where to apply.

Practising the speaking itself

Every part of this talent area that is not a written test is spoken: explaining a build, defending a decision, thinking out loud on an unfamiliar problem. That is the same underlying skill the PSLE English oral assesses, and it responds to practice.

PSLEPrep’s free three-minute diagnostic is built for the PSLE English oral, not for DSA-Sec, and it does not simulate any school’s selection process. It is simply a way to hear how your child sounds answering a question out loud.

Try the free 3-minute speaking diagnostic →

The DSA-Sec Interview Preparation Pack — coming soon

We are still writing it, so leave your email and we will send it the day it is done rather than send you something half-finished. A free guide for parents: how the DSA-Sec selection process works, the kinds of question interviews tend to explore, and how to help your child practise talking about their own interests and work — drawn from MOE’s published DSA-Sec framework.

Question archetypes and the thinking behind them, with worked examples we wrote ourselves.
What the process actually involves, stage by stage, and where to find each school’s own published details.
Early access to our DSA-Sec interview practice tool, expected November–December 2026.

These are not real questions from any school, and we neither publish nor ask for actual interview content. DSA-Sec selection decisions are made entirely by each school, on its own criteria. No preparation — including ours — can guarantee or improve the chance of an offer. PSLEPrep is independent and is not affiliated with, endorsed by, or connected to the Ministry of Education or any school.

Questions parents ask

Science, maths and engineering DSA-Sec: common questions from parents

Will my child be tested, or interviewed, or both?: It depends on the school and often on the specific talent area within it. Published material describes a two-hour written problem-solving paper at one school, selection tests then a camp at another, a selection test then an interview at a third, and one school whose table says shortlisted mathematics, science and robotics applicants are contacted for interviews or trials — the school choosing which. Read each school's own page.
Does my child need competition awards to apply?: Not universally, but some talent areas publish hard gates. One school states that its mathematics talent area considers only Platinum awardees of a particular olympiad round. Other schools publish that applicants without prior experience may apply and will be assessed on potential. These are opposite policies at different schools, which is why the school's own document is the only source worth acting on.
How do panels tell whether a child really did the project?: By questioning specifics — why a design decision was made, what a notebook entry meant, what was tried before this. That is a synthesis published by a preparation provider rather than a first-hand account, so treat it as a well-argued hypothesis. It is also a good description of how anyone would check, and preparing for it costs nothing.
What if the project was a team project?: Say so, clearly, and name your own part without shrinking it or inflating it. Robotics in particular is commonly team-based, and panels know this. The failure mode is a child who uses “we” for everything and cannot say what they personally did.
Are there group tasks in STEM selection?: Yes at some schools, and they are published. One school runs a group STEAM challenge as its second phase; another runs a multi-day selection camp for shortlisted candidates. What happens inside those rooms is almost entirely undocumented — no first-hand account of one exists in the public record, and any provider claiming to simulate the real thing is over-claiming.
Should my child memorise a script for the project talk?: No. Two independent sources with opposing commercial interests — a coach who sells preparation and a parent who had been through selection — both say memorised answers come apart as soon as the question shifts. Practise the material until it is familiar, not until it is fixed.
Is a STEM DSA-Sec place a commitment to a STEM CCA?: Published school documents state that a student admitted through DSA-Sec must take part in activities related to the talent they were selected for from Year 1 to Year 4, cannot take part in S1 Posting, and cannot transfer after PSLE results. For a child whose interest is genuinely wide, four years pointed at one area is worth thinking about before applying.

Science, maths and engineering DSA-Sec interviews: explaining the work, not displaying it