A Punggol student admitted through Posting Group 2 (PG2) can build readiness for G2 Computing by learning how computers take input and produce output, why spreadsheets use cell references and formulas, how Scratch programs control actions and choices, and what it means to protect information online. PG2 is a Secondary 1 admission grouping under Full Subject-Based Banding, not an automatic Computing subject assignment. Effective Computing learning is not just clicking rapidly through software: the student defines a problem, decomposes it into manageable parts, builds a solution, tests it with suitable data, saves the work and explains why it works.

SEAB’s new 2027 G2 Computing syllabus K237 is scheduled for its first examination in 2027. It covers six modules—computer fundamentals, networking, impact of computing, spreadsheets, media software and programming. Graduating candidates have a 90-minute e-examination Paper 1 worth 70 marks and a 2-hour-15-minute lab-based Paper 2 worth 80 marks, each carrying 50% of the overall result. Scratch features in the programming work; MakeCode microcontroller material is assessed only in Paper 1 under the specified syllabus. Families at Edgefield, Greendale, Punggol Secondary and Yusof Ishak Secondary should check whether their own school offers G2 Computing at the relevant year and never assume that all PG2 entrants take K237. This guide focuses on transferable computational thinking and safe, legitimate digital practice.

• PG2 versus G2: admission group differs from an individual Computing subject offer.
• 2027 K237: new G2 Computing syllabus with six modules and separate knowledge and practical papers.
• Computer system: hardware, storage, input, processing and output work together.
• Networks: connect devices and services; users need secure, responsible practices.
• Spreadsheet: structured data, formulas, charts, sorting, filtering and validation.
• Media software: plan graphics, slides and video with clear audience, purpose and attribution.
• Scratch: use events, sequence, variables, selection, iteration, lists and debugging.
• MakeCode: microcontroller concepts are included in Paper 1, not the Paper 2 lab tasks.
• Safety: avoid phishing, protect accounts and respect other people’s privacy and copyright.
• Next step: build one small original Scratch task and verify a separate spreadsheet calculation.

Definition lock: PG2 is not automatic enrolment in Computing

Definition lock: PG2 is not automatic enrolment in Computing

Posting Group 2 determines a route into Secondary 1. Under Full Subject-Based Banding, G2 denotes the level of an individual subject, and different schools offer different later combinations. A pupil may learn general digital skills through school assignments without being enrolled in the specific K237 Computing examination. The subject name should never be inferred solely from the original posting letter.

Parents should obtain the current subject-options booklet and teacher guidance. If Computing is not offered, the teenager can still explore programming and data skills as personal enrichment without pretending that it is compulsory coursework. If K237 is offered at the appropriate stage, the official syllabus becomes useful for planning. Eligibility, timetable and assessment arrangements are decisions to verify with the actual school.

Why the new K237 syllabus matters

Why the new K237 syllabus matters

SEAB explicitly identifies 2027 as the first examination year for G2 Computing K237. This is a new qualifying subject rather than the old G2 mathematics or science syllabus relabelled. Its aims include understanding computer systems, applying computational thinking, using software to create artefacts, understanding cybersecurity and evaluating the impact of technology. It combines conceptual knowledge with practical production.

The year is important: a child starting Secondary 1 in 2027 will not sit the graduating K237 paper that same year. Future syllabuses and school offers may change. Parents should therefore teach enduring skills—logical reasoning, data organisation, media production and safe judgement—rather than promise that one software version or a single list of blocks will stay unchanged forever.

The six syllabus modules each serve a purpose

The six syllabus modules each serve a purpose

Module 1 addresses computer fundamentals; Module 2 networking; Module 3 the impact of computing; Module 4 spreadsheets; Module 5 media software; and Module 6 programming. Together they show that Computing is not limited to writing code. A student also needs to understand devices, communication, responsible use and how digital tools turn information into useful outputs.

Ask which type of question an assignment concerns. A hardware comparison needs different knowledge from a spreadsheet chart. A phishing question is not solved by making a game animation. Strong study plans keep the modules distinct while showing how they interact in genuine projects. A presentation may involve data, a program and ethical media choices within one larger task.

Paper 1 is a theory and application examination

Paper 1 is a theory and application examination

In the 2027 K237 specification, Paper 1 is a 1-hour-30-minute e-examination marked out of seventy. It contains twenty multiple-choice questions worth twenty marks and a structured-question section worth fifty. It covers knowledge, understanding and application across all six modules and contributes half the overall result.

The student needs more than memorised shortcuts. A question may ask why one storage device is appropriate, what a network feature does or why a program gives the wrong output. A good learner explains the relationship and uses evidence from the question. Those skills can be practised using paper scenarios, diagrams and code traces before actual software is involved.

Paper 2 tests creating and saving working files

Paper 2 tests creating and saving working files

Paper 2 is a 2-hour-15-minute lab-based examination worth eighty marks, making up the other half of the overall assessment. It covers practical work in spreadsheets, media software and programming, with approximate task mark allocations of 25, 30 and 25 respectively. Candidates must save the required softcopies, and the allocated time includes file-saving tasks.

This means a creative project that appears correct on screen may still be incomplete if it is saved in the wrong location or format. Practical competence includes following the file instructions and verifying that the completed work is recoverable. Families should use the school’s approved software and working conventions rather than assume a different app behaves identically.

Input, process and output provide a simple model

Input, process and output provide a simple model

A digital system receives input, processes data according to instructions and produces output. In a fictional school attendance calculator, typed numbers are inputs, addition or comparison is the processing, and a displayed total is the output. If the input is wrong, a correctly executed calculation can still produce the wrong real-world information. The system does not know that a typo is a typo without suitable validation.

Ask the teenager to identify the three stages for a familiar program. A barcode scanner, keyboard or microphone supplies input; the processing follows the application; and a display, printed page or sound can be an output. The categories describe roles rather than one fixed device type. A touchscreen can serve as both input and output under different actions.

The CPU and GPU have different responsibilities

The CPU and GPU have different responsibilities

The central processing unit executes many general computational instructions, while a graphics processing unit is designed to perform types of parallel processing useful for graphics and other workloads. K237 includes comparison of integrated and dedicated graphics and memory-sharing arrangements. The student should not assume the GPU replaces the CPU or that a computer with a larger GPU name automatically suits every task.

Ask what workload a particular device must perform. Simple documents and ordinary browsing can have different demands from 3D rendering or video production. Specifications should be compared on relevant criteria rather than advertising slogans. A good answer describes functions and limitations, not just which component sounds more powerful.

Main memory and storage are not synonyms

Main memory and storage are not synonyms

Main memory such as RAM holds working information that programs can access while operating, whereas secondary storage retains files over time in devices such as solid-state drives and hard disks. Closing a program or restarting a machine can affect unsaved work even though properly saved files remain in storage. A student who equates RAM capacity with the total amount of photos saved misunderstands the distinction.

Ask where a program’s working data reside and where a submitted project file is stored. A device with substantial RAM can still have insufficient disk space for large media files. The relevant specification depends on the task. This understanding also supports good habits about saving and backing up work.

Storage speed and capacity answer different needs

Storage speed and capacity answer different needs

Magnetic hard drives and solid-state storage differ in typical access speed, durability characteristics, cost, size and capacity offerings. The K237 syllabus asks learners to compare these properties rather than memorise one universal winner. A faster device is not automatically the cheapest, and a larger capacity does not guarantee the fastest performance.

Give a fictional purchase brief with required file size, portability and budget. Ask the student to select evidence from the specifications and explain one trade-off. This is a computational decision-making exercise, not a product shopping recommendation based on unverified current prices. Good answers match the stated needs.

Bits and bytes carry different units

Bits and bytes carry different units

A bit represents one binary digit, while a byte conventionally contains eight bits in modern computing. Data quantities are described using bytes and larger units such as kilobytes, megabytes, gigabytes and terabytes. Decimal and binary prefixes can differ, so students should follow the definition supplied in the syllabus or question rather than assume all labels use one identical base.

Ask what a listed number measures. Network speed may be quoted in bits per second, while a file size may be quoted in bytes. Eight megabits are not automatically eight megabytes. Unit conversion is part of computational literacy and parallels the care required with centimetres and square centimetres in Mathematics.

Networking lets devices communicate

Networking lets devices communicate

A network connects devices so they can exchange information or share services through suitable infrastructure and protocols. Local networks can operate within a building, while broader networks connect across greater distances. Internet access relies on interconnected networks rather than one giant device in a single building. The student’s role is to understand the connection and its purpose, not to tamper with someone else’s network.

A school may show a diagram of laptops, a switch, a router and access to the internet. Ask which elements connect directly and what route data may take. A router and wireless access point have distinct roles even though a home device may combine functions. Reading the architecture makes troubleshooting descriptions more meaningful.

Wired and wireless communication have trade-offs

Wired and wireless communication have trade-offs

Wired connections use physical media such as cables, while wireless systems transmit through electromagnetic signals. Speed, mobility, interference, distance, security and cost can differ depending on the particular technology and environment. A student should not claim wireless is always faster or wired is always immune to every failure.

Compare two fictional school setups: a fixed desktop lab and a portable device used in multiple rooms. Which connection properties matter for each? The chosen recommendation should follow the stated task. Avoid using the discussion as a reason to connect to unauthorised Wi-Fi or investigate another person’s network.

An internet connection is not a guarantee of accuracy

An internet connection is not a guarantee of accuracy

Network access makes information available but does not verify it. A page can be outdated, misleading or impersonate a legitimate source. The syllabus includes responsible use and awareness of cyber risks because technical connectivity and trustworthy knowledge are different things. A student who can open a website is not automatically able to evaluate its claims.

Ask who published the information, when, for what purpose and what evidence supports it. A device displaying a familiar logo may still be showing an imitation page. Checking an official address and using appropriate trusted channels are part of safe research. This connects Computing with English media literacy and Humanities source evaluation.

Phishing tries to trick rather than break a password mathematically

Phishing tries to trick rather than break a password mathematically

Phishing attempts may use convincing messages or websites to induce someone to disclose credentials, enter payment details or install malicious software. The attack works through deception rather than necessarily by solving encryption. Teenagers should be taught to verify the sender and destination independently, avoid rushed action and report suspicious requests through suitable channels.

An imaginary sample message claiming that a school account will be deleted in five minutes can be discussed without clicking anything. Ask which features create pressure and what safe verification route the student should use. Do not share actual passwords or account recovery codes in a classroom exercise. Security learning must remain protective.

Strong passwords help protect accounts

Strong passwords help protect accounts

Unique, difficult-to-guess passwords or passphrases reduce the chance that an unrelated account breach will expose several services. Multifactor authentication, where available, adds an additional protection. No password choice guarantees absolute safety, and credentials should not be shared casually. The 2027 syllabus also includes understanding precautions such as firewalls, updated anti-malware programs and scam awareness.

Ask the child to describe the practice without revealing any real secret. A fictional password should never be presented as a secure recommendation to copy directly. Learning about safe account management includes recognising when to ask an adult or school IT administrator for authorised support.

Backups protect against more than accidental deletion

Backups protect against more than accidental deletion

Work can be lost through hardware failure, accidental edits, corrupted files, malware or unauthorised access. A backup is an additional recoverable copy stored appropriately, not simply a second shortcut to the same original file. The syllabus includes understanding backups and recovery. A pupil who saves everything to one removable drive has a single point of failure.

Discuss a safe school-approved plan for versioning and backup without uploading personal information to unapproved services. Ask how the student would recover the latest project after a device stopped working. The correct policy depends on school systems and privacy rules. Good digital organisation is part of practical assessment readiness.

Copyright applies to digital projects

Copyright applies to digital projects

Images, video, music, fonts and writing can be protected by copyright even when available online. A student should not assume that anything found through a search engine is free to redistribute. The 2027 Computing syllabus explicitly includes using copyrighted materials responsibly. In a class project, appropriate licence terms, attribution and school policy should be respected.

A simple approach is to use original work or properly licensed resources and record source details. A class project should not paste another creator’s logo or soundtrack without permission merely because it makes the video more polished. Correct media choices are part of computational ethics, not an optional finishing touch.

Privacy settings matter before information is uploaded

Privacy settings matter before information is uploaded

Digital tools may ask for names, location data, images, contact details or access permissions. A teenager should understand the privacy implications before agreeing or publishing. The syllabus emphasises privacy policies, settings and personal information. An ordinary school project rarely needs real classmates’ addresses or medical records to demonstrate a spreadsheet formula or a game score.

Use invented names and fictional data in examples. Ask what information is necessary for the project and whether it can be minimised. If school instructions require particular information, follow approved procedures. Digital literacy includes knowing that a technically functioning project can still be inappropriate if it exposes others’ private data.

A spreadsheet is a model of structured data

A spreadsheet is a model of structured data

Spreadsheets organise information into cells arranged by rows and columns. Each cell has an address such as B3, which combines column B and row three. A collection of related rows can represent records, while columns represent fields such as date, category or amount. Students should distinguish a spreadsheet table from an unstructured page of numbers copied for decoration.

A fictional reading log might have book titles in column A, minutes spent reading in B and dates in C. The column headings determine the meaning of values. Before calculating a total, ask whether every entry uses the same unit and whether blank cells or textual values need attention. Clean data is the foundation for useful formulas.

Formulas must begin with accurate references

Formulas must begin with accurate references

In Excel, an expression such as =B2+C2 adds numeric values from the two named cells. A wrong reference can produce a mathematically valid result from the wrong inputs. The computer will not always know the intended human meaning. A student must check that references correspond to the requested quantities, not merely that a result appears.

With fictional data B2=15 and C2=20, the formula gives 35. But if C2 contains a fee belonging to another row, the arithmetic is no longer answering the correct question. Computational correctness includes input selection, not just symbol syntax.

Relative and absolute references behave differently

Relative and absolute references behave differently

Copying a formula containing B2 to another row can change the relative reference to B3. An absolute reference such as $D$1 can remain fixed when copied, according to normal spreadsheet conventions. This is useful for calculations with one shared rate or constant. A student who fails to understand this distinction may create a whole column of incorrect outputs after dragging a formula.

Imagine a fictional worksheet with quantities in column B and a single conversion rate in D1. The formula =B2*$D$1 preserves the shared rate when filled down. The numbers should remain hypothetical and not be presented as live Singapore currency rates. The purpose is understanding reference behaviour.

A formula output can look plausible while being wrong

A formula output can look plausible while being wrong

Suppose a spreadsheet is meant to sum five scores but the formula accidentally includes a header or skips the last row. The result may still resemble an ordinary total. This is why debugging a spreadsheet requires checking the selected range and comparing with a small manual calculation. A neat chart cannot repair an incorrect source formula.

Ask the student to identify the range represented by =SUM(B2:B6). Then verify each included cell against the intended question. An off-by-one error is common in both spreadsheets and programming. Learning to inspect boundaries is one of the most useful transferable computational habits.

Functions solve recurring calculation problems

Functions solve recurring calculation problems

Spreadsheets provide functions for operations such as totals, averages, counts, comparisons and lookups. The K237 syllabus includes familiar functions and more advanced spreadsheet tools at the specified graduating level. Parents should match practice to the teacher’s current lessons rather than insist that every Secondary 1 pupil memorise an extensive function catalogue.

A student can start with =SUM(B2:B5) and =AVERAGE(B2:B5), checking what each function does and whether blank or nonnumeric values matter. Then move to the particular syllabus functions assigned in class. A formula is useful only when its range and interpretation fit the underlying question.

Charts communicate quantities, not just colours

Charts communicate quantities, not just colours

A spreadsheet chart translates selected data into a visual representation, but an incorrect category range, scale or unit can mislead. A bar chart may suit comparisons among groups, while a line chart can show a sequence of values across time. Students should choose a graph based on the message and data structure rather than the most decorative chart type.

Ask what the horizontal and vertical variables represent and whether a zero baseline is important for the intended comparison. A chart of invented Punggol reading counts must be labelled as a fictional practice dataset. A graph is not evidence of a real local pattern unless its data and source are trustworthy.

Sorting and filtering answer different questions

Sorting and filtering answer different questions

Sorting changes the display order of records based on a chosen criterion, while filtering shows only those that meet specified conditions. A learner might sort fictional book loans from largest to smallest or filter records for one category. Neither operation should silently alter the meaning of individual rows or cause values to separate from their correct labels.

Ask the student to keep each record intact during sorting. A classic error is sorting one column alone while leaving other columns fixed, misaligning data. A reliable worksheet requires selecting the appropriate table or range. This is data integrity, not merely the convenience of a button.

Data validation can prevent avoidable input errors

Data validation can prevent avoidable input errors

Validation applies rules to incoming data, such as allowing whole numbers only within a selected range. The K237 syllabus includes setting criteria and displaying custom error messages. A fictional reading log might accept a nonnegative number of minutes, while a category field might use an approved list. Such constraints can catch some accidental entries before they enter a calculation.

Validation is not the same as proving truth. A number inside an allowed range might still be fabricated or recorded for the wrong person. Ask what errors the rule can prevent and what errors remain possible. A strong student understands both the purpose and limitation of data-validation tools.

Media work begins with audience and purpose

Media work begins with audience and purpose

Module 5 involves media software, including computer graphics, slide presentations and videos. The first step should be deciding what the audience needs to understand or do, not choosing a flashy animation. A slide intended to teach two scientific ideas may need one clear diagram and a short explanation rather than fifteen unrelated visual effects.

Ask the teenager what problem the media artefact solves. A well-structured storyboard can identify the message, relevant images, text and sequence before any software is opened. This planning prevents wasted effort and ensures the final file can be evaluated against the task’s actual requirements.

Vector graphics differ from raster images

Vector graphics differ from raster images

Vector graphics describe shapes through geometric objects such as paths, nodes, lines and curves, while raster images store pixel-based information. Vector elements can often be resized cleanly without the same pixelation associated with enlarging a low-resolution raster image. The syllabus includes constructing and editing vector objects. Students should not assume that every picture file remains sharp at unlimited magnification.

A simple original school logo made of vector shapes can illustrate the difference without copying a protected brand mark. Ask what happens to line edges when the design is resized and what editing controls remain available. The media choice should follow the intended use, not the assumption that one file format is always best.

A storyboard clarifies a video before editing

A storyboard clarifies a video before editing

A storyboard sequences scenes with notes about image, spoken or written content, timing and transitions. The K237 syllabus includes producing presentations and videos through media software. A pupil who starts importing random clips without a clear sequence may produce an attractive but confusing result. Planning is part of technical competence.

Imagine a three-scene original explainer about why a spreadsheet formula must be checked. Scene one identifies the problem, scene two demonstrates a wrong reference and scene three shows the corrected output. The storyboard ensures a logical educational message. No expensive editing software or copyrighted music is needed to understand the method.

Frames and frame rate affect video

Frames and frame rate affect video

A video is formed from sequential frames displayed over time. A higher frame rate can create smoother motion for suitable footage but may increase data and processing requirements. The actual effect depends on compression, resolution, encoding and content, so one number does not describe complete quality. This is part of the K237 media-software awareness.

Ask the student whether they are describing the number of frames per second or the length of the finished video. These variables are different. Export settings should follow the school’s instructions. A file that looks smooth but is in the wrong required format may still be unsuitable for a practical submission.

Scratch is a visual programming environment

Scratch is a visual programming environment

Scratch allows learners to build programs by assembling graphical instruction blocks. It can control sprites, respond to events, change variables, use conditions and loops and create simple interactive games. The goal is computational thinking: the order and logic of instructions matter even when the programmer does not type conventional text code.

Ask what the program should accomplish before selecting blocks. A simple task might move a sprite when a particular key is pressed and display a response under a defined condition. The student should describe the intended behaviour in plain language, build the necessary logic and test it. Dragging colourful blocks without understanding their effect is not programming mastery.

Sequence controls order of execution

Sequence controls order of execution

A program executes instructions according to its control structure. If a sprite must first reset to a starting position and then move forward, reversing those steps may produce different behaviour. A beginner may blame the software when the real issue is an instruction placed in the wrong order. The code may be syntactically valid but logically unsuitable.

Ask the pupil to trace the blocks from the triggering event and predict the result before running them. This dry run helps develop mental execution skills. A good programmer can explain what each step changes. The same idea applies to spreadsheet workflows and everyday procedures: sequence matters when later operations depend on earlier states.

Variables store values that can change

Variables store values that can change

A variable may hold a score, count, timer or other program state. A Scratch game can initialise score to zero, then increase it when an appropriate event occurs. If a new round begins without resetting the score, the program may carry the previous value accidentally. This is an example of a state-management bug, not a computer hardware failure.

Ask which block creates or updates the variable and which event should trigger it. A useful check is to run the program twice and compare the starting value. Programmers should think about initialization, updates and expected output. A variable name should describe its purpose so the code remains understandable.

Conditions allow a program to choose

Conditions allow a program to choose

An if instruction executes an action when a condition is true; an if–else structure offers alternative paths. A simple original game might award a point when a sprite touches a designated target but do nothing otherwise. The condition must match the intended rule. If the wrong comparison is used, the program may award points at the wrong time.

Ask the child to state the condition in plain English and predict outcomes for both true and false cases. The K237 syllabus includes relational and logical operators and limited nested conditions. Beginners should master one clear decision before stacking many branches. A short test table can reveal missing cases.

Loops repeat instructions with a purpose

Loops repeat instructions with a purpose

A loop repeats a group of instructions according to a count or condition. Scratch includes repeat, forever and conditional loop patterns. A program that repeats without an appropriate stopping or waiting mechanism can behave unexpectedly, while a loop with an incorrect bound may perform one step too many or too few. Understanding the condition is as important as recognising the block shape.

Imagine a sprite required to move five steps of ten units each. A repeat-five loop can express the action efficiently. Ask what changes if the repeat count becomes four. The child should predict the output before running it. This demonstrates abstraction and reduces mechanical copying of identical blocks.

An event tells code when to begin

An event tells code when to begin

Many Scratch scripts begin when a key is pressed, a sprite is clicked or a message is received. The event is the trigger that starts the associated instructions. A learner who constructs a correct sequence without any trigger may wonder why nothing happens. The issue is not the arithmetic or movement blocks but how the script is initiated.

Ask the student to identify the trigger and test it deliberately. Different events can start different scripts, potentially interacting through shared variables or messages. The program’s behaviour should be reasoned about as a system rather than as one long list of coloured shapes.

Lists organise several related items

Lists organise several related items

The K237 syllabus includes list or array operations such as creating, reading, inserting, deleting and checking for items. A Scratch list can store several fictional quiz questions or high scores. The programmer needs to know which entry is being accessed and how the list changes when an item is inserted or removed. Indices are not just decorative numbers.

Ask what happens to later positions when an early entry is inserted. A program that reads the wrong index can return a plausible but incorrect answer. Use small imaginary lists and trace each operation by hand. The student can develop the concept safely without collecting classmates’ actual personal data.

Custom blocks help decompose a problem

Custom blocks help decompose a problem

Long programs become easier to manage when repeated or complex behaviour is represented through named custom blocks or functions. For example, an original game might have one block to reset the round and another to update the score display. Breaking work into logical components is decomposition, one of the syllabus’s computational-thinking aims.

Ask the learner what one block promises to do and what input or state it depends on. Clear names and distinct responsibilities make testing easier. A custom block should simplify the program’s structure rather than merely hide a collection of unrelated actions. Good design helps another reader understand the solution.

Debugging is evidence-based correction

Debugging is evidence-based correction

Debugging begins by identifying the difference between expected and actual behaviour. The programmer then inspects the relevant sequence, values and conditions, makes a specific correction and tests again. A sprite that moves too far may have an extra repeat; a score that never changes may be using the wrong event or variable. Guessing and replacing every block at once makes the cause harder to learn.

Encourage a short bug record: expected output, observed output, likely faulty step, change made and retest result. This mirrors the error-log approach used in Mathematics. A learner who can explain the reason a fix worked has developed transferable reasoning, rather than simply arriving at a visually pleasing final screen.

MakeCode has a different assessment role

MakeCode has a different assessment role

The K237 syllabus includes microcontroller material and MakeCode blocks for Paper 1 knowledge and application, while the lab-based Paper 2 practical programming tasks use the specified Scratch scope. SEAB’s syllabus tables explicitly distinguish Scratch blocks for Papers 1 and 2 from MakeCode blocks for Paper 1 only. Students should not assume every microcontroller exercise belongs in the practical examination.

A family can discuss input from sensors and output to displays using official diagrams without buying or rewiring hardware. If the school offers authorised microcontroller activities, follow its equipment and safety instructions. The important learning goal is understanding how a digital system senses and responds, not collecting gadgets.

A fictional Scratch quiz illustrates complete thinking

A fictional Scratch quiz illustrates complete thinking

Imagine a small original quiz game that asks a fictional arithmetic question, checks an answer and adds one point if it is correct. First define inputs and expected outputs. Then initialise score, wait for a response, compare it with the intended answer and update the state. Test a correct input, an incorrect input and an empty response if relevant to the design.

This is a hypothetical teaching project and not a ready-made official exam task. Ask the learner what each test checks and which blocks would be involved. The game does not need elaborate graphics to demonstrate decomposition, variables, events and conditions. A small correct program is more educational than a large buggy one copied from an online tutorial.

A fictional spreadsheet shows how data errors spread

A fictional spreadsheet shows how data errors spread

Imagine a made-up club attendance table with counts 12, 15, 9 and 14. The expected total is fifty. If a SUM formula accidentally skips the last row, it returns thirty-six. The function is doing exactly what it was instructed to do, but the selected range is wrong. The resulting error can also flow into a chart or a percentage calculation.

Ask the pupil to inspect the range, compare the result with a manual sum and revise the formula. The numbers are invented and do not describe actual Punggol club attendance. This exercise shows that checking the meaning of inputs matters as much as knowing the software command.

Data verification should precede chart design

Data verification should precede chart design

A chart generated from incorrect or misleading data may look polished and persuasive. Before designing it, check that each column is correctly labelled, the numbers represent the intended population and period, and missing values have been handled appropriately. A student who adds animation to a wrong graph has improved appearance without improving accuracy.

Ask whether the source is authentic or hypothetical. Real school data require permission and appropriate privacy safeguards. For an ordinary tutorial, fictional data are sufficient. Clear labels and a defensible data source are part of professional-quality computational work.

Practical files need sensible names and formats

Practical files need sensible names and formats

A project can contain an excellent spreadsheet or Scratch program but be difficult to assess if the required file was never saved. K237’s lab examination explicitly includes saving softcopies within the allocated time. Students should use the correct naming and location specified by their school, verify the file exists and reopen it when permitted to check integrity.

Do not rely only on an automatically generated temporary file or a screenshot when the task requests editable source material. An exported video is not automatically the same as the original editing project. File management should be practised as part of the work, not postponed until the final minute.

Design a small test plan instead of clicking randomly

Design a small test plan instead of clicking randomly

A good test plan includes typical inputs, boundary cases and invalid inputs. A quiz with scores from zero to ten should be tested at zero and ten as well as a middle value. A spreadsheet field restricted to nonnegative counts should reject an entry of minus one according to its validation rule. These tests reveal mistakes that normal cases can miss.

Ask the child what should happen before running each test. Then record the actual result and whether it meets the requirement. Testing is evidence, not a ritual. This habit transfers to Mathematics checking, Science fair tests and software quality work.

AI output still needs human verification

AI output still needs human verification

Some students may use an AI assistant to generate code or text. Such output can be incorrect, misleading or unsuitable for the task, and school rules may restrict its use. The learner remains responsible for understanding the work they submit. A Scratch script that the child cannot explain is not evidence of mastery merely because it runs once.

Use authorised assistance for feedback, then have the teenager explain every variable, condition and test. Copyright, privacy and source attribution still apply. Tools can support learning without replacing independent reasoning. The teacher’s policy governs what kind of assistance is allowed in school assignments and assessments.

A Punggol school project should not publish personal records

A Punggol school project should not publish personal records

A map of Punggol or a fictitious school data table can be a useful project setting, but creating a public file with classmates’ names, phone numbers, attendance, grades or medical details would be inappropriate without proper authority. For practice, use imaginary names and clearly labelled sample records. The lesson does not require live personal data.

Ask whether each field is necessary and who may access the completed file. Good computing practice minimises unnecessary collection and applies the school’s storage and sharing controls. A technically correct spreadsheet can still be unsafe or unethical if it exposes private information.

Devices and software differ across schools

Devices and software differ across schools

SEAB identifies approved software and examination infrastructure requirements for K237, but an individual school’s teaching environment, equipment and subject offers need confirmation. A parent should not assume that buying a particular laptop or software subscription is necessary merely because the subject exists nationally. Some school facilities may already provide relevant access.

Check the current official school requirements before making purchases. A basic computer with appropriate authorised software can be sufficient for much conceptual practice, and paper-based tracing helps where a device is unavailable. The immediate priority is the student’s learning goal, not collecting expensive hardware.

A twelve-day skills sequence for an eligible learner

A twelve-day skills sequence for an eligible learner

Start with input, process, output, hardware and storage. Then explore networking and account safety, including scams, backups and privacy. Next build a small spreadsheet table and verify formulas, cell references and a chart. Later plan a short presentation or video with proper attribution. In the second week practise a Scratch script with a variable, condition, loop and event, followed by deliberate debugging and saving.

This is an optional readiness sequence for a child whose school supports the appropriate learning, not a mandatory Secondary 1 syllabus. If the pupil cannot yet understand a cell reference, simplify the spreadsheet task before introducing lookup functions. If a program fails, trace the logic before adding features. Progress should be demonstrated independently.

How parents should interpret Computing feedback

How parents should interpret Computing feedback

Computing mistakes have different causes: misunderstanding a network concept, choosing the wrong spreadsheet range, omitting a Scratch event, using an inappropriate chart or saving the wrong file. The tutor should locate the first invalid step and explain its effect. Repeatedly telling the teenager to ‘be more careful’ without identifying the error offers little practical guidance.

Ask the learner to reproduce a small task after the correction, without the model answer visible. They should be able to explain why a formula calculates the intended value and why a Scratch script follows its planned branch. A working artefact and a sound explanation together provide stronger evidence of learning than an attractive finished screenshot alone.

Official sources, related guides and your next step

Official sources, related guides and your next step

SEAB’s 2027 G2 Computing syllabus K237 confirms the six modules, the two examination papers, Scratch and MakeCode scope, spreadsheets, media and responsible computing. The 2027 G2 subject index identifies Computing as new. MOE’s Full Subject-Based Banding guidance explains why PG2 admission does not automatically specify later subject offerings.

For hands-on design thinking rather than Computing assessment, see the PG2 Design & Technology guide. For digital homework and school-device responsibilities, use the PG2 SLS and Personal Learning Devices guide. Your next step is to ask the school whether Computing is offered at the relevant level, then have your teenager explain one input–process–output example, check a fictional spreadsheet formula and build a small original Scratch script while keeping personal data private.

Frequently asked questions about PG2 G2 Computing and K237

Does PG2 automatically mean the student can sit K237 Computing?

No. PG2 concerns admission; the school’s actual subject offers and eligibility must be confirmed.

Is G2 Computing new in the 2027 SEC?

Yes. K237 is identified by SEAB as a new G2 Computing examination with first examination in 2027.

What are the six modules?

Computer fundamentals, networking, impact of computing, spreadsheets, media software and programming.

How much is Paper 1 worth?

Seventy marks in a 90-minute e-examination, contributing 50% of the overall syllabus assessment.

What is Paper 2?

A 2-hour-15-minute lab-based paper worth eighty marks and 50% overall, assessing practical spreadsheets, media and programming.

Does MakeCode appear in Paper 2 practical programming?

No. The specified MakeCode microcontroller blocks are examined in Paper 1 only.

Why does a spreadsheet need validation?

To catch some unsuitable or out-of-range entries before they affect calculations, though it cannot prove all entries are truthful.

What is Scratch debugging?

Testing expected versus actual behaviour, identifying the incorrect instruction or state and retesting after correction.

Can a pupil use real classmates’ data in a practice spreadsheet?

Use fictional data unless the school has proper authorisation and privacy controls.

Does a new Secondary 1 student in 2027 take that year’s K237 SEC paper?

No. That SEC syllabus describes candidates graduating in 2027, not new entrants following a later course.

Editorial note, 6 November 2026: K237 is the 2027 graduating G2 Computing syllabus. Actual school offers, approved software and examination arrangements should be confirmed with the assigned school. All project data, games and learner situations described as hypothetical are teaching examples, not real personal records or published school results.