G3 Science tutorials for Chinatown students should develop more than fast recall. A student needs to know which model applies, which measurements matter and how far the evidence allows a conclusion to go. At eduKateSG, our three-student small-group approach makes those decisions visible through worked examples, individual explanation and independent application.
For families comparing G3 Science tuition, a Physics, Chemistry or Biology tutor, or SEC Science preparation, the appropriate plan begins with the student’s actual course. Lower-secondary integrated Science, upper-secondary combined Science and separate sciences should not be treated as identical programmes with different amounts of homework.
This page is for families travelling from Chinatown; it does not represent a local branch or an affiliation with a school bearing the Outram name. Consultations and suitable placements are arranged at 8 Fourth Avenue, Singapore 268674, near Sixth Avenue MRT. Confirm the student’s year, subject combination, current topics and timetable before choosing a class.
Our aim is a learner who can approach an unfamiliar question with a defensible first step. That may be a diagram, a defined quantity, a conservation relationship or a carefully limited claim. The lesson should teach the student how to choose that step, not only how to follow it once the tutor has chosen it.
Ask about G3 Science class suitability on WhatsApp or arrange a parent–student consultation.
Science Learning Lens for Chinatown — G3
Science in a Dense Heritage District
Chinatown offers familiar prompts for scientific questions about shade, thermal conditions, food, materials, light and human movement. We do not treat the neighbourhood as a measured experiment. Instead, we convert everyday observations into stated paper scenarios where quantities, variables and evidence are clearly defined.
Shade, Heat and Surface Comparisons
A learner may notice that sheltered and unsheltered places feel different, but Science asks which quantity would be measured and what else must be controlled. In a hypothetical comparison, two surfaces are measured at the same stated time under controlled conditions. Students distinguish surface temperature from air temperature and avoid claiming more than the supplied readings support.
Food Contexts Without Turning Science Into Cooking Advice
Food gives accessible contexts for changes of state, energy transfer, dissolving, concentration and biological processes. We use safe paper-based scenarios and supplied data rather than instructing students to perform unsupervised experiments. The scientific task is to identify the process, variable or evidence that the question actually targets.
Built Materials and Property–Function Reasoning
A dense older-and-newer built environment invites questions about transparency, strength, conduction, insulation and flexibility. We provide a table of material properties and a defined design requirement, then ask students to justify a choice. The learner practises transferable property–function reasoning instead of memorising one familiar material example.
Crowd Data as a Graph-Reading Exercise
A fictional dataset may show pedestrian counts across several time intervals. Students first read axes and scale, then describe the pattern, compare intervals and identify anomalies. Only after that do they discuss possible explanations. A pattern in supplied data is not automatically proof of a cause.
From Chinatown Context Back to General Science
The local context is removed after the first successful explanation. Numbers, diagrams and wording change while the underlying concept remains. The learner must recognise the same relationship without the place name. This final transfer step keeps local context useful without making the learning locally dependent.
A More Important Transition Than It First Appears
A learner can know a chapter well enough to recognise its familiar questions and still have difficulty deciding when its ideas apply. The transition to stronger G3 performance is therefore not only an increase in content. It is an increase in responsibility for selecting, qualifying and connecting that content.
During a worked example, the teacher has already selected the relevant model. In an unfamiliar problem, the student may have to distinguish a constant value from a constant rate, identify a limiting condition or decide whether a graph supports a causal claim. Those decisions belong to the answer even when they are not written as separate steps.
We make them explicit during teaching. Before calculating, a student explains why the relationship is suitable. Before concluding, the learner identifies the evidence and any necessary assumption. Before comparing, the student checks whether the measurements refer to equivalent conditions.
The Hidden Science Problem: An Answer Can Be Correct Under the Wrong Assumption
Consider an original paper exercise involving a moving object. Its speed increases uniformly from 2 to 10 metres per second over four seconds. A student calculates an acceleration of 2 metres per second squared. Under the stated uniform-acceleration condition, that is correct.
Now change the question. The same starting and ending speeds are recorded, but the speed does not change uniformly. The same division gives the average acceleration over the interval; it does not establish the acceleration at every instant. A familiar calculation remains useful, but the claim attached to it must change.
Adrian, Jo and Ben are fictional teaching characters, not student testimonials. Adrian completes the arithmetic quickly and overlooks the condition. Jo notices the change but abandons a still-useful average calculation. Ben waits for a new formula. Each needs a different clarification about what the available information supports.
This is the kind of distinction we want a G3 student to handle calmly. An assumption is not decorative wording. It defines the situation in which a model or calculation can be used.
Why Consider Three-Student Science Tutorials?
A class of three allows a useful discussion without removing individual accountability. Every student first commits to an answer and a reason. The tutor can then compare the reasons rather than allow the fastest speaker to supply the group’s thinking.
One learner may select the right equation but use the wrong interval. Another may read the interval correctly but confuse gradient with area. A third may calculate correctly and then describe the wrong physical quantity. These are different errors hidden beneath apparently similar numerical work.
The class format is valuable when it creates time to examine those differences and test a fresh attempt. Small size alone does not establish quality. The lesson must still require the learner to explain, decide, calculate and revise without being led through every step.
G3 Science Under Full Subject-Based Banding
G3 is a subject level, not a synonym for Secondary 3. The student’s year and subject level are separate. MOE’s Full Subject-Based Banding guidance explains the framework, while the secondary syllabus directory provides the relevant curriculum routes.
For the 2027 SEC, G3 includes separate Physics K323, Chemistry K324 and Biology K325. It also includes combined Science in Physics–Chemistry K326, Physics–Biology K327 and Chemistry–Biology K328. G3 therefore does not automatically mean taking all three separate sciences. See SEAB’s school-candidate subject list.
At lower secondary, our priority is the student’s current integrated programme. At upper secondary, the exact combination and individual subject syllabuses guide the depth and assessment preparation. We do not assume that the same paper structure applies to combined and separate Science.
What We Work On in G3 Science Tutorials
Three different readings from one motion graph
Return to the original example in which speed increases uniformly from 2 to 10 metres per second over four seconds, with motion in one direction. The ending speed is 10 metres per second. The acceleration is the gradient, (10 − 2) ÷ 4 = 2 metres per second squared. The distance is the area under the speed–time graph: one half × (2 + 10) × 4 = 24 metres.
These are three different quantities extracted from the same graph. A student who treats every graph question as find the gradient will miss the distinction. We ask what the vertical coordinate, slope and area represent before deciding which calculation answers the question.
Next, the learner explains why the trapezium calculation was suitable. The straight-line change in speed makes the average of the two endpoint speeds appropriate over the interval. If the graph’s shape changes, the area must be found from the new shape or information supplied.
The goal is not to memorise three answers. It is to connect the requested quantity with the relevant feature of the representation.
Resultant force rather than the largest visible force
In a hypothetical horizontal-motion problem, a 5-kilogram object experiences a 14-newton force to the right and a 4-newton resistive force to the left. Assume these are the relevant horizontal forces and that the vertical forces balance.
The horizontal resultant is 10 newtons to the right. Using resultant force = mass × acceleration gives an acceleration of 2 metres per second squared to the right. Dividing 14 by 5 would use the applied force while ignoring the opposing interaction.
We then ask what would happen to the calculation if the two horizontal forces were equal. A zero resultant does not, by itself, say that the object is stationary. The initial motion matters. This follow-up separates the idea of balanced forces from the everyday assumption that no net force means no movement.
Turning effects and the correct distance
Suppose an original lever problem gives a perpendicular force of 12 newtons acting 0.25 metre from a pivot. Its moment magnitude is 3 newton metres. The distance is measured perpendicularly from the pivot to the force’s line of action, not automatically along any convenient edge in the drawing.
We change the diagram while preserving that perpendicular distance. The learner should recognise that a longer decorative arm does not necessarily change the relevant moment. Then we change the line of action and ask what must be reconsidered.
This type of exercise teaches the student to read geometry as part of the Physics. A labelled measurement is not always the measurement needed in the relationship.
Circuit analysis with an explicit boundary
For a paper-based ideal-circuit example, two resistors of 4 ohms and 8 ohms are connected in series across 12 volts. Ignore internal resistance and assume the resistor values remain constant. The total resistance is 12 ohms and the current is 1 ampere.
The potential differences across the resistors are then 4 volts and 8 volts. They add to the supply value. A learner who assigns 12 volts independently to each series resistor has not identified the relationship between the components and the source.
We redraw the connection as parallel and ask students to restart the reasoning, not reuse the series total. Changing a connection changes the model. These are written exercises, not instructions to alter household wiring or test circuits without appropriate supervision.
Chemical equations that distinguish ratios from masses
In an original stoichiometry exercise for a relevant G3 course, use the reaction 2Mg + O₂ → 2MgO. Take the relative atomic masses as Mg = 24 and O = 16, and assume 4.8 grams of magnesium reacts completely with sufficient oxygen to form magnesium oxide only.
The magnesium amount is 4.8 ÷ 24 = 0.20 mole. The equation gives equal mole amounts of magnesium used and magnesium oxide formed, so the product amount is 0.20 mole. Its molar mass is 40 grams per mole, giving a product mass of 8.0 grams.
The product is heavier than the initial magnesium because oxygen has been included. A student who expects equal masses because the coefficients are equal has confused a mole ratio with a mass ratio. The correction returns to what the coefficients count.
We explicitly identify the assumptions: sufficient oxygen, complete reaction and the stated product. This is a calculation exercise, not a recommendation to burn magnesium at home. The scientific purpose is to connect conservation, chemical identity and quantitative reasoning.
Reaction rate and final amount as separate claims
Imagine two invented gas-volume graphs. Curve A reaches 40 cubic centimetres sooner than curve B, but both eventually level at 50 cubic centimetres under the stated measurement conditions. The first conclusion concerns the rate of gas production over relevant intervals; the second concerns the final collected volume.
A faster reaction does not automatically mean a larger final amount. Students should identify which graph feature supports each claim: the slope or time taken for a chosen volume, compared with the final plateau.
We then ask whether the collected volume necessarily equals all the gas produced. A leak or another collection limitation would affect that interpretation. The student learns to separate the ideal experimental model from possible limitations in an actual setup.
Biological comparisons with different starting sizes
Consider original teaching data for two samples. Sample A increases in mass from 4.0 to 4.4 grams. Sample B increases from 6.0 to 6.4 grams. Both gain 0.4 gram, but their percentage changes differ: A increases by 10%, while B increases by about 6.7%.
This does not mean percentage change magically makes every experiment fair. Other conditions still matter. It does show why the choice of comparison must be explained when starting quantities differ.
In an appropriate osmosis question, the student would also need to interpret the biological process and measurement method. Were samples handled consistently before weighing? Was the relevant tissue and duration controlled? The numerical comparison and biological explanation must support one another.
Explaining across biological scales
A question can begin with an organism and require an explanation involving organs, cells or molecules. The learner must decide which scale supplies the missing mechanism. Naming an organ is not always sufficient, but writing everything known about the organism is not a substitute for choosing the relevant process.
For example, an explanation involving increased activity may need to connect an organism’s demand with cellular energy release and the transport of required substances. The exact chain depends on the question and the syllabus, not on a universal sentence template.
We ask Clara to draw arrows between the levels of her explanation. Ethan checks whether each arrow states a mechanism or merely places two facts next to each other. This makes a missing causal link easier to identify than a general instruction to add detail.
Testing the limits of a conclusion
Suppose an invented investigation compares a biological process at three temperatures and the middle condition produces the highest measured rate. The evidence identifies the best-performing condition among those tested. It does not locate an exact optimum across every possible temperature.
A useful next investigation might test smaller intervals near the apparent maximum while maintaining relevant controls. The recommendation follows from the uncertainty in the result. It is more informative than a generic suggestion to use better equipment.
We want students to be confident enough to state a limited conclusion. Scientific precision sometimes means narrowing a claim rather than making the answer sound more certain.
Our First-Principles Teaching Method
We first identify the learner’s independent starting point. A short question may ask for the model rather than a complete solution. What quantity does the graph’s area represent? Which substance is moving? Which condition makes the equation applicable? These questions expose the decisions that longer solutions can conceal.
Next, we rebuild the earliest unstable link. If a reaction calculation fails because a coefficient is being read as a mass ratio, more arithmetic practice is not the first remedy. If a graph calculation fails because area and gradient are confused, the physical meanings need to be separated.
We then keep a clear boundary around the concept. An ideal circuit, a single uniform-motion interval or a reaction with stated assumptions makes the first relationship visible. Complexity increases only after the student can explain the simpler situation.
Supported practice follows the explanation. The next question removes some support. A later question changes the representation or asks for the reasoning in reverse. We are not satisfied with a correct answer that can only be reproduced when every feature resembles the original example.
Finally, we return after a delay. The student may need to recognise the same principle among unrelated questions. This check tests selection and recall together. It gives the tutor a more useful basis for planning than the number of pages completed during the first lesson.
What a Focused Lesson Can Look Like
An illustrative 90-minute sequence begins with ten minutes of delayed retrieval and fifteen minutes of concept reconstruction. The next twenty minutes use supported practice, followed by twenty minutes of independent work in a changed representation.
Fifteen minutes then introduce a short mixed or timed task. The final ten minutes compare the original and corrected reasoning, identify a remaining uncertainty and set a focused continuation task. The timing is an example of lesson organisation, not a published guarantee about every Science placement.
For a secure learner, more time can be spent on unfamiliar applications. For a learner with a prerequisite gap, more time may be needed before independent work. The non-negotiable educational feature is that the student produces and explains work; a lesson should not be entirely a performance by the tutor.
Three G3 Science Learning Pathways
Repair: We find the concept that is disrupting several later tasks. A weak understanding of ratios can affect chemical calculations; a weak distinction between gradient and area can affect graph interpretation. The repair is tied to current schoolwork so the student can see why it matters.
Stabilisation: We test whether the learner can maintain accuracy as representations and topics change. Practice includes explaining why an attractive wrong approach fails, not only producing the correct approach. The objective is dependable selection rather than recognition alone.
Extension: We deepen the reasoning through model limitations, alternative explanations and unfamiliar data. An advanced task should make the student think more precisely, not merely carry out a longer calculation using a formula from a different course.
A Worked Evaluation: What Does a Repeat Actually Improve?
Suppose a class measures the time required for a process. Repeated trials produce slightly different values because judging the endpoint is difficult. A student suggests repeating the trials and reporting an average. That may help summarise the variation, but it does not automatically resolve a consistently late endpoint judgement.
We ask the learner to distinguish variation between trials from a shared bias in the method. The proposed improvement should identify which limitation it addresses. A more objective endpoint rule, where practical, may address a different problem from simply increasing the number of trials.
The next task asks the student to describe a limitation that repetition would not solve. This reverses the usual question and tests whether the learner understands the purpose of the improvement rather than remembering a phrase frequently found in mark schemes.
A Worked Interpretation: Interpolation Is Not Unlimited Prediction
Imagine an original dataset with measurements at inputs of 2, 4, 6 and 8 units. A question asks for an estimate near 5 units. Another asks for a prediction at 80 units. Both involve a relationship, but the second moves far beyond the observed range.
The student should not treat a straight line drawn through a short range as proof that the same pattern continues indefinitely. The physical or biological system may introduce a limit, a change in mechanism or a condition not represented in the original observations.
We ask what assumption would be required to extend the pattern and what additional evidence could test it. This is useful preparation for unfamiliar data questions because it links numerical reasoning with the limits of the model.
How We Reduce Repeated Mistakes
A useful correction distinguishes a wrong model from poor execution of a correct one. A student who identifies the right force balance and then makes an arithmetic slip needs a different intervention from a student who never considered the opposing force.
We also separate a scientifically false answer from an answer that is true but does not address the question. Describing a graph accurately is not the same as explaining its mechanism. Stating a mechanism may not answer a question asking for evidence from the supplied data.
The correction record can remain compact: the first failed decision, the repaired principle, a fresh application and a later check. A record that only preserves the model answer does not reveal whether the learner can now choose the right method.
Teaching Ahead Without Rushing
A calm first encounter with an upcoming concept can be useful. We may introduce the relevant quantities, establish the central diagram or clarify the language before schoolwork becomes more demanding.
Pre-teaching should not be used to conceal an unstable foundation. If the student cannot distinguish a ratio of moles from a ratio of masses, a harder reaction problem will not repair that distinction. If a learner cannot interpret a simple motion graph, more complicated graphs may simply multiply the confusion.
The decision to move ahead should follow evidence from independent work. A completed chapter is not the same as an understood chapter.
Combined Science Practical Preparation Must Be Planned
The 2027 G3 combined Science scheme includes a one-hour multiple-choice paper worth 20%, two 75-minute discipline papers worth 32.5% each, and a 90-minute practical test worth 15%. See SEAB’s combined Science scheme. These figures should not be copied onto a separate-science revision plan.
Our planning response is to distinguish written understanding from practical execution. A student may explain a measurement method accurately on paper and still need supervised practice setting up equipment, recording readings promptly and responding when a result is unexpected.
Apparatus diagrams, data interpretation and method evaluation can support preparation, but they do not replace every part of laboratory experience. Families should ask exactly which practical arrangements a proposed class includes. School laboratory work and the student’s actual examination requirements remain important.
For separate Physics, Chemistry or Biology, use the individual syllabus linked from the official G3 directory. The subject title is not enough to determine paper timing, weighting or practical requirements.
What Progress Should Look Like
We look for more controlled decisions. The student states an assumption when it matters, chooses the graph feature that matches the requested quantity, keeps chemical ratios attached to the right entities and writes explanations that connect evidence with mechanism.
A further sign is better self-correction. The learner can identify why an earlier method was unsuitable and explain what must change in the next attempt. This is different from recognising a tutor’s correction once it has been supplied.
We compare like with like: independent attempts with independent attempts, and timed work with similarly timed work. No fixed grade or improvement timeline is guaranteed. The aim is a clearer record of what the learner can do and a teaching plan that responds to that record.
When Should an Chinatown Student Begin?
Consider a consultation when the student’s effort is substantial but the error pattern remains unchanged. The learner may repeatedly choose a familiar formula for an unsuitable situation, write long explanations with a missing causal link or lose control when a question changes its representation.
A strong student may seek a more demanding environment for unfamiliar problems. A struggling student may need a narrower repair. A student who is already learning independently and managing the workload may not need another class. The decision should be based on a specific purpose, not a general fear of falling behind.
Access From Chinatown and Class Details
The teaching venue is at 8 Fourth Avenue, Singapore 268674, near Sixth Avenue MRT. Use SBS Transit’s Sixth Avenue station information for the arrival end and plan the route from the student’s actual starting point.
Consider the complete weekday rather than only the rail journey. School dismissal, walking, a meal, the lesson and the trip home all affect whether a slot is sustainable. We do not claim an identical travel time for every Chinatown household.
Format: Three-student small-group tuition. Focus: G3 Science at the actual secondary year and subject combination. Materials: Worked explanations, selected applications, schoolwork review and purposeful continuation tasks. Placement: Subject to readiness, topic compatibility and availability. Confirm duration, fees, timetable and practical arrangements directly before enrolment.
What Parents Can Bring to the Consultation
Bring recent marked assessments, a current worksheet, practical feedback and the school’s topic sequence where available. For upper-secondary students, include the exact Science combination or separate subjects and the examination year.
An uncorrected attempt is useful because it preserves the learner’s original decision. We ask what help was given, which feedback was understood and whether the same issue appeared again. We also discuss the actual time available between lessons so the plan remains manageable.
Frequently Asked Questions
Does G3 mean taking three separate sciences?
No. The official G3 subject list includes both combined Science pairings and separate Physics, Chemistry and Biology. Confirm the student’s actual course through the school and SEAB’s subject directory.
Will more difficult questions always help?
Only when the difficulty is purposeful. A harder question can test a secure concept in an unfamiliar setting. It can also obscure a basic gap beneath more complicated arithmetic. We choose the next task according to the decision that needs developing.
Should students learn every model answer by heart?
Model answers can demonstrate precise language, but the learner should understand which evidence and mechanism each sentence expresses. We change the context and ask the student to adapt the explanation rather than reproduce wording that no longer fits.
How do you help a student who calculates accurately but explains poorly?
We ask the learner to interpret each quantity and justify the relationship before writing the final explanation. A diagram or short oral account can reveal whether the issue is understanding, selection or expression. The response should match the cause.
Can a student focus on one separate Science?
The consultation should establish the exact subject and support needed. Suitable placement depends on the available timetable and class configuration. An article covering the Science pathways is not a promise that every subject has an immediate vacancy.
Does written practical preparation replace laboratory work?
No. Written tasks can strengthen planning and interpretation, but equipment handling and real observations require suitable supervised experience. Confirm the actual practical provision and continue to take school laboratory work seriously.
What should parents look for besides a test score?
Look for independent starts, relevant explanations and fewer repeated errors in changed contexts. Ask the student to explain one correction without reading the model answer. That conversation can show whether the learner understands the decision that improved.
Helpful Reading for Chinatown Families
PSLE Science Tuition | Chinatown · A Student’s Life | Chinatown · Education and Tuition | Chinatown · G3 Science Tutorials | Outram Park · G3 Science Tutorials | Raffles Place
G3 Science Tutorials for Chinatown Families
A strong G3 learner does not need every question to look familiar. The student can identify what is known, select a suitable model, state an important assumption and build an explanation that remains within the evidence.
For students who are behind, we repair the missing connection. For students who are inconsistent, we check whether control survives changed representations and time. For students who are ready, we deepen the reasoning without confusing unnecessary complexity with understanding.
Science Learning Blueprint for Chinatown — G3
Reading the City as a Science Text
A student travelling through Chinatown moves through changing shade, road surfaces, sheltered spaces, moving vehicles, indoor cooling and outdoor heat. We use such ordinary observations only as prompts for scientific questions: what quantity would describe the change, what could be measured, what alternative explanation must be controlled, and what evidence would justify a conclusion? The place is not treated as a laboratory result. It is a source of observable questions that can then be converted into safe paper-based models.
From Crowded Information to the Relevant Variable
Dense urban environments contain many simultaneous changes. That makes them useful metaphors for scientific selection. A question may provide five pieces of information while only two control the calculation. We teach the learner to mark the requested quantity, circle the evidence that bears on it and deliberately leave irrelevant information unused. The habit is especially valuable in data-response questions, where the difficulty often comes from selection rather than arithmetic.
Heat, Shade and Surfaces
A simple urban heat question can compare two hypothetical surfaces placed under the same stated conditions. The student first distinguishes surface temperature from air temperature, then identifies what was actually measured and whether the comparison was controlled. We avoid claiming that a particular Chinatown location has a measured temperature unless data are supplied. The teaching point is how to build a defensible comparison from stated evidence.
Movement, Time and Rate
Urban movement gives a familiar context for rates without requiring claims about actual journey times. If a model object covers 120 metres in 80 seconds, the average speed is 1.5 metres per second. A second journey may cover a greater distance yet have a lower average speed if it takes proportionally longer. Students learn to compare ratios rather than the largest raw number.
Evidence Before Explanation
We sometimes present a fictional Chinatown-style urban dataset with noise level, temperature or footfall as abstract values. The learner first describes the pattern, then identifies which additional evidence would be needed for an explanation. This trains an important scientific boundary: a correlation in supplied data is not automatically a proven cause.
Systems Thinking in a Dense District
Science becomes easier when the learner sees systems rather than isolated facts. A transport system contains inputs, constraints and flows; a biological system contains structures, materials and processes; an electrical circuit contains connected components. We use the city only as an analogy for organisation, then return immediately to the exact scientific system named in the syllabus question.
G3 Science Requires Model Boundaries
At G3, students are expected to know not only a relationship but the conditions under which it can be used. A calculation may be correct under one assumption and overclaimed under another. We teach learners to state important assumptions, interpret the result and distinguish an average across an interval from a value that applies at every instant.
Physics, Chemistry and Biology Keep Their Own Logic
Physics often asks which system and quantity matter. Chemistry asks students to connect observations, particles and symbols. Biology frequently requires a causal chain across structures and processes. We use one learning architecture—diagnose, explain, practise, retrieve, transfer—without pretending the disciplines are interchangeable.
Combined and Separate Science Need Different Depth
For the 2027 SEC, G3 includes separate Physics K323, Chemistry K324 and Biology K325, and combined Science pairings K326, K327 and K328. The student’s actual school combination determines content depth and assessment preparation. A generic G3 worksheet programme cannot substitute for that alignment.
Diagnose the First Unstable Decision
A wrong answer is the end of a chain, not the diagnosis. We ask where the chain first changed direction. Did the learner misread the graph, choose the wrong quantity, recall the wrong concept, omit a condition, reverse a relationship or explain with language that was too broad? The repair begins at that point. This prevents a student who needs one conceptual distinction from being assigned an indiscriminate stack of questions.
Arrange a Parent–Student Consultation
Tell us the student’s secondary year, Science combination or separate subject, current topic and one recurring difficulty. That gives the consultation a practical starting point.
Arrange a G3 Science consultation on WhatsApp
eduKateSG
8 Fourth Avenue, Singapore 268674
Near Sixth Avenue MRT
Three-student small-group tuition
By appointment