For the 2027 Singapore–Cambridge Secondary Education Certificate, G2 Science is offered in three combinations: K223 Physics and Chemistry, K224 Physics and Biology, and K225 Chemistry and Biology. Start with the combination on the student’s school records. A guide covering all three disciplines does not mean that every candidate studies or sits examinations in all three.

A useful learning plan connects knowledge, calculations, explanations and experimental reasoning. It should help the student recognise what a question is asking, select an appropriate idea and check the answer independently. This guide distinguishes official examination arrangements from our original teaching examples. The numerical examples and practice scenarios below are illustrations, not SEAB questions, predicted questions or an official marking scheme.

• Check the subject code: K223, K224 and K225 identify different two-discipline combinations.
• Prepare for the correct papers: four paper components are taken across two science sessions.
• Develop more than recall: explanations, unfamiliar information and experimental reasoning matter.
• Practise selectively: use the examples relevant to the student’s actual combination.
• Diagnose errors: distinguish missing knowledge, misreading, calculation mistakes and unsupported conclusions.
• Verify arrangements: current SEAB notices and school instructions govern the examination.

Which combination does the student actually take?

The SEAB G2 school-candidate directory for 2027 maps K223 to Science (Physics, Chemistry), K224 to Science (Physics, Biology), and K225 to Science (Chemistry, Biology). These are combinations within the Science qualification, not three interchangeable names for an identical course.

Write the correct code at the front of the revision folder and check it when choosing resources. A worksheet can teach useful science while still being unsuitable as a rehearsal for the student’s actual assessment. Avoid accumulating materials simply because their covers say “secondary science”.

For a family using this guide, read the shared learning sections and the examples relevant to the two disciplines being studied. The third discipline may be interesting, but it should not silently displace the student’s assigned work. Confirm any uncertainty with the school rather than inferring the combination from a textbook left on the desk.

The official paper structure

The 2027 combined G2 Science syllabus specifies four components per candidate: K223 takes Papers 1–4; K224 takes 1, 2, 5 and 6; K225 takes 3–6. Physics uses Papers 1/2, Chemistry 3/4, and Biology 5/6.

Each discipline’s pair shares a 75-minute session: a 20-mark multiple-choice component and a 30-mark structured component, contributing 20% and 30% respectively. The syllabus advises no more than 30 minutes for the multiple-choice component. Structured Section A carries 22 compulsory marks; Section B offers two eight-mark questions, of which one is answered.

The assessment-objective balance is approximately 45% knowledge with understanding, 45% information handling and problem-solving, and 10% experimental skills and investigations. The scheme has no separate laboratory practical examination, but practical understanding is assessed within the theory papers. Therefore, “no separate practical paper” does not mean “no need to understand experiments”.

Turn the paper structure into a sensible study plan

The important planning consequence is that neither science discipline should be treated as an optional extra. A learner may enjoy one more than the other, but a revision schedule built entirely around the preferred subject leaves a predictable weakness. Begin by looking at recent work from both.

Within each discipline, distinguish recognising an answer from constructing one. A student can select a correct option yet struggle to explain the underlying relationship in writing. Another can explain an idea well but overlook a small condition in a multiple-choice question. Those are different skills to practise.

Our suggested planning sheet has three columns: what the student can do independently, what needs a prompt, and what is not yet understood. Put concrete tasks in the columns rather than whole-subject labels. “Can calculate a speed but forgets to convert minutes to seconds” is much more useful than “Physics is weak”.

Begin diagnosis with one piece of real work

Choose a recent question the student answered incorrectly. Before showing a model answer, ask what they thought the question required. Let them describe the first decision they made. This can reveal whether the difficulty began before any scientific calculation or explanation.

A learner who selected the wrong quantity needs help interpreting the task. A learner who selected the correct relationship but entered a number incorrectly needs a checking routine. A learner who copied a correct sentence without understanding it needs a smaller conceptual example. Prescribing the same extra worksheet to all three misses the distinction.

Record the first incorrect step and one follow-up task. Do not use the exercise to assemble a list of faults. The purpose is to identify a teachable point, demonstrate it clearly and check whether the learner can use it in a slightly different situation.

A Physics example: calculate average speed with units

For an original practice question, suppose an object travels 150 metres in 30 seconds. Average speed is total distance divided by elapsed time, so the calculation is 150 ÷ 30 = 5 metres per second. The relationship is explained in OpenStax’s motion equations. Our numbers are invented for this example.

Now change only the time statement: the object travels 150 metres in half a minute. The physical situation is unchanged, but the student must recognise that half a minute is 30 seconds before calculating a result in metres per second. A correct formula is not enough if the inputs use incompatible units.

Ask the learner to predict how the answer should change if the same distance takes longer. That prediction provides a useful reasonableness check. If a slower journey produces a larger calculated average speed, inspect the interpretation and arithmetic before accepting the result merely because it came from a calculator.

A second Physics example: changing velocity

Suppose an object’s velocity changes from 2 metres per second to 10 metres per second in the same direction over four seconds. The change is 8 metres per second, giving an average acceleration of 2 metres per second squared. See OpenStax’s explanation of acceleration for the underlying distinction between velocity and its rate of change.

A useful incorrect response is 10 ÷ 4. Ask what that calculation leaves out. The student should identify the starting velocity rather than simply memorise that the answer is wrong. The correction concerns the meaning of “change”, not an arbitrary subtraction rule.

Then change the starting value while keeping the final value and time fixed. Ask for a prediction before calculation. The activity helps the learner connect a mathematical operation to the situation it describes. Do not introduce extra motion conditions that the question has not supplied.

A third Physics example: use the resultant force

For a one-dimensional invented problem, let a four-kilogram object have a ten-newton force to the right and a two-newton force to the left. The resultant is eight newtons to the right, so its acceleration is two metres per second squared to the right. The relevant relationship is described in OpenStax’s Newton’s second law section.

Ask the learner why dividing ten by four would not answer this problem. The missing step is to account for the opposing force. Drawing two labelled arrows can make that omission visible without turning the exercise into a complicated diagram.

Next, ask what can and cannot be concluded from a resultant force of zero. It does not, by itself, prove that an object is stationary. The student needs to distinguish a statement about changing motion from a statement about the current motion. That contrast is a useful diagnostic conversation, not a trick to embarrass the learner.

Read a graph before choosing a calculation

Our graph-reading exercise begins with invented data: at times zero, two and four seconds, the distances recorded are zero, six and twelve metres. Ask the student to identify the quantities and units before interpreting the pattern. On a distance–time graph, these points lie on a straight line with a gradient of three metres per second.

Now imagine the vertical axis represents a different quantity. The same visual shape would no longer justify the same physical interpretation. The lesson is to read the axes rather than memorise that a particular line shape always means a particular motion.

For practice, ask the learner to describe what the data show in words, then support the description with a calculation. A statement such as “it goes up” is too vague to establish what is increasing. The student should identify the measured quantity and the interval being discussed.

A Chemistry example: count atoms rather than altering formulas

For a desk-based equation exercise, consider hydrogen and oxygen forming water: 2H₂ + O₂ → 2H₂O. Counting gives four hydrogen atoms and two oxygen atoms on each side. The balancing principle is explained in OpenStax’s chemical-equation chapter. This is a symbolic exercise, not an instruction to perform the reaction.

Ask the student what would happen if they changed the formula of water to make the numbers look convenient. They would be changing the substance represented rather than correctly balancing the equation. The useful distinction is between a coefficient in front of a formula and a subscript within it.

Have the learner create a small atom-count table and check both sides. When a correction is needed, change one coefficient and recount. This gives a repeatable method that is more dependable than visually guessing a combination of numbers and stopping when the line looks balanced.

Separate an observation from an interpretation

Use a fictional written report: “Bubbles appeared in a liquid, and the container’s measured mass decreased.” Ask which words describe observations and which additional claims would require evidence. Naming a particular gas without a stated test would go beyond the information supplied.

The student can consider possible explanations without presenting one as established fact. For example, a question may provide information about material leaving a system. The answer should connect that information to the observation rather than declaring that matter has simply disappeared.

Our teaching suggestion is to use three sentences: what was observed, what explanation is supported, and what information is still needed. Not every examination answer needs three sentences; this is a practice scaffold. It helps the learner resist the temptation to fill an evidential gap with a familiar but unproven chemical story.

Use particle representations carefully

A drawing can help a student think, but it can also hide ambiguity. For an invented activity, provide a key saying that one symbol represents one particle of type A and another represents type B. Ask the learner to describe what the diagram establishes before attaching chemical names to the symbols.

Change the diagram and ask what changed: the number of particles, their arrangement, the grouping of symbols or the space represented? The answer should refer to the key and the stated model. Do not assume that a larger drawn circle automatically means a particle has gained mass or that a missing symbol proves material was destroyed.

This exercise develops careful use of representations without requiring a hazardous experiment. When moving back to school work, compare the learner’s explanation with the model actually taught. A decorative drawing is less useful than a simple representation with a clear key and an explanation of its limitations.

A Biology example: distinguish diffusion from osmosis

Diffusion concerns net movement down a concentration gradient; osmosis concerns water moving through a selectively permeable membrane under the relevant conditions. The distinction is explained in OpenStax’s passive-transport chapter. The source covers more detail than this school-level illustration; it is not an additional SEC syllabus.

For a paper exercise, draw two compartments separated by a membrane that allows water through but not the illustrated solute. State that the temperature and initial pressure are equal. Ask the student to predict the initial net movement of water when one side contains a more concentrated solution. They should identify the moving substance and the membrane condition, not write only “particles move”.

Keep the practice diagram explicit. An answer depends on what can cross the membrane and the conditions stated. The purpose is to improve reasoning and vocabulary, not to encourage unsupervised experiments with biological material or to replace the school’s practical teaching.

A Biology-style data exercise: calculate a percentage change

Suppose a fictional investigation records an initial sample mass of 5.0 grams and a later mass of 5.5 grams. The increase is 0.5 grams. Relative to the initial mass, the percentage increase is 0.5 ÷ 5.0 × 100 = 10%. These are invented numbers for a calculation exercise, not research findings.

Ask why the denominator is the initial mass in this calculation. Then compare a second sample that increases from 10.0 to 10.5 grams. The absolute increase is again 0.5 grams, but the percentage increase is 5%. Equal changes in grams need not represent equal proportional changes.

Do not infer the biological cause from these numbers alone. A question must provide the relevant conditions before the learner can justify a scientific explanation. This separates the mathematical conclusion, which the data support, from a biological interpretation that may require additional information.

Choose a conclusion that matches the evidence

Imagine a supplied table showing a larger measured response in condition B than in condition A. A cautious first conclusion describes that comparison. It does not automatically establish that B is better in every situation, that a particular mechanism has been proven or that the pattern will continue outside the measured range.

Ask the student to identify the boundary of the claim. How many conditions were tested? Were measurements repeated? Does the question provide a reason to treat the groups as comparable? These are questions to ask of a supplied investigation, not a demand to invent missing observations.

A useful correction changes the wording as well as the answer. “In the supplied results, B had a larger measured response” is more defensible than “B always causes the largest response”. The learner is practising precision about evidence, not learning to avoid conclusions altogether.

Explain variables through the question being investigated

For an original planning discussion, imagine comparing how long identical objects take to travel a fixed distance under two stated conditions. Ask what the investigator deliberately changes, what is measured and what should remain comparable. The specific choices must follow the question, not a memorised list of “temperature, time and amount” inserted into every answer.

If two important conditions change together, ask whether the result can confidently be attributed to only one. This reveals why a fair comparison needs thought before data are collected. The exercise can be completed with a written scenario; it does not require families to construct equipment.

For school investigations, follow the teacher’s method and safety instructions. A proposed improvement should name the problem it addresses. “Make it more accurate” is an aim, not a complete change to a method. The student should explain what they would alter and why that alteration would help answer the question.

Use repeated measurements honestly

Consider fictional time measurements of 9.8, 10.1 and 10.0 seconds. Their arithmetic mean is approximately 9.97 seconds. Before presenting a final value, the learner should consider the precision appropriate to the task rather than copying every calculator digit without thought.

Now suppose a further reading is very different. Do not teach the student to delete it merely because it makes the pattern less tidy. Ask whether the scenario gives a reason to identify an error, whether the method should be checked and whether additional evidence is available. An unexpected result is a question to investigate, not permission to manufacture a preferred answer.

Our suggested practice record keeps the original readings visible alongside any calculation. That makes the reasoning inspectable. A neat final mean with no trace of the underlying data can hide a transcription error or an unjustified exclusion.

Check scales, tables and units systematically

In a printed or digital diagram, begin with the labelled scale. Count the intervals between labelled values before assigning a value to a small division. A student who assumes every division is one unit may produce a confident but incorrect reading.

For a table, keep the measured quantity and unit clear in the heading. Ask the learner to explain what one row represents. A set of numbers without labels may be impossible for another reader to interpret, even when the arithmetic is correct.

Our checking routine has three passes: confirm the values copied from the question, confirm the operation performed and confirm how the result is reported. This is a suggested study habit, not an official marking rule. It helps locate the type of error instead of reducing every wrong number to “carelessness”.

Respond to the instruction, not a remembered paragraph

Read the command in the question before writing. A request to describe a pattern differs from a request to explain it. A calculation needs a relationship and the appropriate inputs; an inference needs a connection between the evidence and the conclusion.

For practice, give the learner the same invented table with two different questions. First ask for the largest recorded value. Then ask for an explanation of a stated relationship using additional information supplied in the question. The second task should not receive the first answer merely expanded into a longer sentence.

The official syllabus includes a glossary that families can consult. Use it to clarify expectations, but do not assume a single memorised sentence structure answers every question beginning with the same command. The content and evidence of the particular task still matter.

Build explanations from a specific observation

Our suggested explanation scaffold begins with the fact from the question that needs explaining. Next identify the relevant scientific relationship. Finally connect the relationship to that fact. This is a teaching framework, not a promise that three sentences always receive full marks.

In the force example, a useful explanation refers to the two opposing forces and the resultant before discussing acceleration. A vague statement that “the object moves because of force” leaves out the relationship that distinguishes this problem from another one.

Ask the learner to remove any sentence that does not help answer the question. More writing is not automatically more reasoning. A precise response may be short, while a long answer may contain contradictions. The goal is a clear, supported connection that another reader can follow.

Learn from multiple-choice errors

After an incorrect multiple-choice response, do not stop at circling the correct letter. Ask which assumption made the chosen option look attractive. Did the student miss a unit, confuse two terms, overlook a condition or select a result from an incomplete calculation?

Our correction method is to write one sentence explaining why the selected option fails and one sentence explaining the relevant principle. Then attempt a new question that changes the surface details. Repeating the same letter from memory does not demonstrate that the original misconception has been repaired.

Do not assume every distractor deserves an elaborate analysis. Focus on the actual error and the idea it exposes. The learner’s time is better spent correcting a specific misunderstanding than copying an entire answer key into a notebook.

Approach structured questions in parts

Read the full question to understand the situation, then identify what each part asks. A later part may require using information already established, but the learner should not assume every answer is independent or that every sub-question calls for the same type of response.

For a calculation, show enough working to make the selected relationship and substitution clear. For an explanation, refer to the relevant information rather than rewriting the whole scenario. For a choice question, follow the paper’s instructions carefully instead of spending time answering every alternative.

In practice, compare an incomplete response with an improved version. Ask what was added: a missing unit, a justified relationship, a relevant comparison or a clearer conclusion. That comparison makes improvement tangible without pretending that a private tutor can guarantee how an unseen national-examination response will be marked.

Use practical learning safely

Practical understanding includes recognising what a method measures, how observations are recorded and how a conclusion is supported. The absence of a separate laboratory examination should not lead families to replace all investigation work with memorised definitions.

At home, use supplied diagrams, fictional datasets and teacher-approved activities. Do not recreate chemical reactions, work on mains electricity, handle unknown substances or use biological materials merely because an online exercise mentions them. School laboratory work should follow the teacher’s instructions and the school’s safety arrangements.

When discussing a method, ask the student to identify the purpose of each step. A learner who can explain a procedure is better placed to notice when a question changes a condition. Copying a laboratory description word for word may conceal uncertainty about what the procedure is intended to establish.

Create a correction record that remains useful

A useful correction entry contains the original question reference, the first incorrect idea, the corrected relationship and a new check. Keep it brief enough to revisit. “Remember to revise harder” does not tell the student what to do differently.

For the speed example, an entry might say: “I used minutes while reporting metres per second. Convert the elapsed time before substitution.” For the equation example: “I changed a subscript instead of adjusting coefficients. Count each element on both sides.” These are different errors and deserve different reminders.

Revisit the idea after a delay using new numbers or a changed context. The aim is to establish whether the correction can be used without the earlier answer in view. A beautifully maintained notebook is not, by itself, evidence that the underlying misconception has gone.

Plan a week around both disciplines and one clear priority

Begin with the school’s current work. Select one main difficulty in each discipline and one previously taught skill to revisit. For example, a K223 student might work on interpreting a motion graph and checking an equation, while also practising unit conversions.

Keep the tasks small enough that the learner can explain the result. One session can focus on rebuilding an idea, another on applying it to a new question, and another on reviewing independent work. This is an example routine, not a prescribed SEAB timetable or a guarantee about learning speed.

Allow the plan to change when evidence changes. A student who can now perform a task independently does not need endless identical practice. A student who remains confused may need a different explanation or a prerequisite repaired before adding time pressure.

Introduce timing after understanding is stable

First establish that the learner can carry out the method accurately without an artificial rush. Then use a short timed task to identify where time is being lost. The problem may be slow retrieval, repeated rereading, uncertain choices or unnecessary rewriting.

Compare the timed work with an untimed attempt. If the same misconception appears in both, the main need is conceptual teaching rather than a faster pace. If the untimed response is sound but the timed version omits essential steps, practise a manageable response routine.

For examination rehearsal, follow the actual paper structure and school guidance. Include reading and checking within the session. Do not teach a home timing rule as though it overrides the candidate instructions or guarantees an ideal allocation for every learner.

Ask the school questions that can be answered

Bring an actual question or work sample to a discussion. Ask which part of the response shows understanding and where the reasoning first breaks down. A specific example is more useful than asking whether the child is generally “good at Science”.

Confirm the subject combination, materials and upcoming learning priorities. For a possible change in subject level or combination, ask about the school’s actual options, readiness criteria and timetable. This guide does not create an entitlement to a different course or establish a future admissions outcome.

Where motivation or attendance is becoming difficult, describe what has been observed without assigning blame. A learner may need help understanding instructions, rebuilding a foundation or managing the workload. The next intervention should respond to the evidence rather than assume that every difficulty is caused by insufficient effort.

Frequently asked questions

Does every G2 Science student take Physics, Chemistry and Biology?

No. The three subject codes identify three different two-discipline combinations. Check the school record and use the matching papers. The presence of all three disciplines in a shared syllabus document does not make all three compulsory for every candidate.

Does no separate practical examination mean practical reasoning is unimportant?

No. The assessment objectives include experimental skills and investigations. Students should understand methods, observations and data, while following appropriate safety arrangements for practical learning.

Should a student simply memorise more definitions?

Definitions can support understanding, but a useful study plan also checks whether the student can interpret information, use relationships and explain a conclusion. The worked examples in this guide are intended to make those additional demands visible.

Can a correct numerical answer still hide a weakness?

Yes. Ask the student to explain the method and attempt a changed example. A remembered answer or a lucky selection may look correct without establishing independent understanding.

Are these examples official examination questions?

No. They are original teaching illustrations. Use the school’s materials and official SEAB resources for the relevant examination requirements, and do not treat this article as a prediction of future questions.

The next useful step

Select one recent response from each of the student’s two disciplines. Ask the learner to explain the question, the method and the check they used. Record one precise improvement to make in each, then choose a follow-up task that tests that improvement.

Subject codes and assessment arrangements checked against the 2027 SEAB G2 Science syllabus on 15 November 2026. This independent guide provides selected teaching examples, not an exhaustive replacement for the syllabus. Current official notices and school instructions take precedence.