A number can be accurate and still leave a listener confused. Someone may say that a solution measured 0.1288 mol/dm³, that a percentage rose by 4.76%, or that a test was repeated three times. Each statement contains information. Yet the audience may not know what was measured, what the figure should be compared with or how much confidence to place in it. Clear spoken English turns numbers into meaning without making the evidence sound stronger than it is.
This learner’s guide addresses a narrow, practical difficulty: how to explain quantitative information aloud when you are not handing the examiner a table to read. It is not another guide to generic persuasion, posture or a scripted personal opinion. It follows Guide 0410 on chemical amounts and titration calculations and Guide 0411 on numerical precision. Those volumes teach the technical calculation. This volume teaches the distinct English skill of making the conclusion useful to a listener while retaining its units, reference, scope and uncertainty.
The official SEAB index for 2027 G3 syllabuses, last updated on 26 October 2026, lists G3 English Language as K300. The K300 English syllabus describes four papers. Writing accounts for 35% (70 marks, one hour 50 minutes); Comprehension for 35% (50 marks, one hour 50 minutes); Listening for 10% (30 marks, about 45 minutes); and Oral Communication for 20% (30 marks, about 20 minutes, with 10 minutes of preparation). Oral Communication includes a planned response to a video stimulus and spoken interaction. These facts describe the official examination, not a promise that a particular video or statistical topic will appear.
You are not expected to give a chemistry lecture in a G3 English oral examination. The examples here deliberately cross subjects so that you can practise a transferable speech habit. If the actual stimulus concerns public spaces, technology, volunteering or healthy routines, the same habit helps: identify the point, choose the relevant detail, give its meaning and keep the claim within what you know.
A listener cannot see the workings inside your head
When you read a data table, you can return to a column heading, check an axis and inspect a footnote. A listener hears your words once as they move past. Speech therefore needs a different order from written working. Begin with the conclusion the listener should understand, supply the most useful number, tell them what it refers to, and then state any important limitation.
Consider a fictional community project. A display says ’40 participants out of 50 completed a survey.’ A speaker who says only ‘forty, which is eighty percent’ has omitted the denominator and the event. A better sentence is: ‘Of the fifty people who took part in this survey, forty completed it, which is eighty percent.’ The ratio, the population and the activity travel together.
This is not about making speech formal. It is about preventing avoidable questions. Who were the fifty? What did the forty do? Is the figure a count or a percentage? The more simply you answer these questions, the easier it is for a listener to follow your next idea.
Speak the reference before interpreting the number
The word ‘higher’ is incomplete until you name what one thing is higher than. The same applies to ‘faster’, ‘cheaper’, ‘more efficient’ and ‘safer’. These words are relational. If the relation is missing, a confident sentence may carry almost no useful information.
Suppose a practice worksheet gives two fictional event plans costing $88.80 and $97.20. It is accurate to say the first plan costs $8.40 less for the specified number of stations. It is not enough to say ‘it saves eight dollars’ if the audience needs to know the unit and comparison. It is worse to say ‘it is the best plan’ because a cost difference cannot settle questions of suitability, staffing or safety.
In a spoken answer, put the reference next to the claim: ‘For six stations, Plan A costs eight dollars and forty cents less than Plan B on the stated items.’ That sentence is short, but it avoids three errors. It names the two plans, the shared basis and the restricted cost category.
What to say about a percentage
A percentage always needs a whole. ‘Fifteen percent more’ usually means relative to an earlier amount, but the speaker should state that earlier amount when it matters. A rise from eighty dollars to ninety-two dollars is a fifteen-percent increase on the original eighty. The reverse journey requires a smaller percentage decrease because the new reference is ninety-two.
You do not have to speak out every fraction in an oral response. Give the meaningful comparison and, when needed, one supporting detail: ‘The cost went from eighty to ninety-two dollars, a fifteen-percent increase on the original price.’ The listener understands both the change and the base.
A percentage point is different. If a participation rate changes from 40% to 50%, the rate rises by ten percentage points; the relative increase is twenty-five percent of the original 40%. Confusing the two changes the claim. Where the distinction matters, the words ‘percentage points’ should be heard clearly, not hidden in a rushed sentence.
Do not read a data table aloud
Imagine a table with six measurements from a fictional school demonstration. Reading every value in order sounds thorough, but an examiner cannot easily build a picture of what matters. The speaker should select a useful relationship, not turn speech into a list of decimals.
A better approach is to choose one anchor result, one comparison and one conclusion. For example: ‘The first two readings were close, but the third was noticeably higher. That means I would be careful about presenting a single average without explaining how it was chosen.’ The language preserves an important observation without drowning the listener in digits.
If exact figures are needed, give only the figures necessary to support the point. You might add ‘The first two differed by just five hundredths of a cubic centimetre.’ A curious examiner can then ask about the third reading. You have created a path for conversation rather than a wall of numerical details.
Distinguish observations from explanations
A figure may describe what happened without explaining why it happened. Suppose a chart shows more people attending a workshop after the organisers changed its timing. It does not automatically prove that the new timing caused the increase. Other things might have changed: outreach, weather, the group surveyed or the event itself.
In speech, this uncertainty is easy to manage with one clear sentence: ‘Attendance was higher after the change, but this comparison alone cannot show whether the timing caused it.’ That is not evasive. It gives the listener an accurate account of the evidence and its limit.
The same principle applies to a laboratory demonstration. A set of close readings may show consistent measurement, but not the absence of every systematic error. The speaker should say ‘The readings were consistent’ rather than ‘The method was completely accurate’. Strong claims require stronger support.
Make technical units sound human
Units help a listener understand the scale and kind of quantity. You can say ‘twenty-five cubic centimetres’, ‘metres per second’ or ‘moles per cubic decimetre’, but you do not need to recite the abbreviation letter by letter unless doing so clarifies an unfamiliar term. If the unit is obscure to your audience, explain its role in a few words.
For example, ‘The concentration was approximately 0.129 moles per cubic decimetre, which describes the amount of substance in a litre of solution under our calculation.’ This sentence is more useful than speaking ‘zero point one two nine m-o-l slash d-m-three’. It translates notation into the concept behind it.
Do not assume that rounding a number for speech gives permission to change its meaning. ‘About 0.13’ can be a sensible spoken approximation when the exact result is 0.1288, but ‘about one’ is not. Be deliberate about how much detail the conversation needs.
Use the order point, evidence, meaning, limit
For numerical claims, a four-part spoken structure works well: make one point, give a useful piece of evidence, explain what that evidence means, and identify a significant limit if one exists. This is a communication pattern rather than an official marking template.
For a fictional school exhibition, a response might begin: ‘I think the demonstration could help visitors see why science needs careful measurement. For example, two readings can be very close while a third is different. That encourages students to ask how results are selected instead of assuming one tidy average proves everything. But the experiment would still need proper supervision and a clear explanation of its limitations.’
The listener now knows the opinion, the example, its meaning and the boundary. More importantly, each sentence advances the thought. There is no need to pile on several unrelated opinions to sound intelligent.
Begin with a view, not a miniature dissertation
A planned response should answer the actual question. If you are asked whether a demonstration is useful, start with your judgement, not a definition of science or a history of measurement. If you are asked what makes information trustworthy, begin with a criterion such as accuracy, transparency or appropriate context.
A useful opening is specific enough to guide the response: ‘I think the display would be useful if it shows visitors how a measurement leads to a conclusion and why results still need checking.’ That opening promises an answer that can be developed with examples.
Avoid trying to sound definitive at all costs. A conditional view may be the best view when it honestly responds to the stimulus. ‘It depends’ is weak on its own, but ‘It depends on whether the organisers explain the figures and offer an opportunity to ask questions’ gives an intelligible criterion.
Speak in meaningful groups
Not every sentence should be the same length. An important number can be followed by a short explanation. A contrast can be marked with a brief pause. Too much speed compresses the units and turns qualifications into afterthoughts.
For example, consider: ‘The figure is eighty percent. That means forty out of the fifty people surveyed completed the task. It does not tell us what the people outside this group would have done.’ The first sentence announces. The second interprets. The third limits. Each has a clear job.
You are allowed to think briefly before a difficult answer. A small pause is preferable to producing an elaborate sentence with several corrections. The goal is listener comprehension, not maximum words per minute. This is compatible with the earlier guide on spoken-interaction repair, which explores a different problem: adapting when follow-up questions change the position you must address.
When numbers are not needed
A number is useful only if it advances the answer. If a question asks why volunteering matters to you, a borrowed statistic about volunteers worldwide may sound impressive but make your answer less personal and less credible. You can speak about experience, values, practical examples and trade-offs without producing a figure.
In a data-based question, however, a number may be crucial to distinguish two choices. Ask: ‘Would omitting this number make my claim unclear, untestable or unfair?’ If not, keep the answer simple. If yes, include the minimum figure necessary to support the point.
Do not invent precision for fluency. Saying ‘Most people in Singapore agree’ without a trustworthy source is not stronger than saying ‘Some people may value this approach because…’ The latter is measured and makes no fake statistical claim.
Explain differences before declaring winners
A speaker describing two groups should ensure the groups are comparable. If one group contains fifty students and another contains ten, raw totals can be misleading. You may need a percentage or a rate. But even comparable rates require attention to time, definitions and sampling.
Imagine twenty of fifty Group A students and six of ten Group B students completed a task. A has more completions in absolute number; B has a higher completion rate. The two statements are both true. Neither justifies declaring one group more capable on the basis of a single exercise.
In a strong spoken response, make the distinction explicit: ‘Group A had more completed tasks, but Group B had a larger proportion of its members completing the task.’ That sentence shows command of the evidence without going beyond it.
Use caution as information rather than apology
Words such as ‘approximately’, ‘under these conditions’, ‘among the surveyed group’ and ‘based on these readings’ protect important distinctions. They need not make your speech timid. They tell the listener exactly where a conclusion applies.
A weak attempt sounds apologetic: ‘I’m not sure, perhaps, maybe everything is possible.’ A stronger attempt sounds grounded: ‘Based on the readings provided, the second trial appears more consistent, although we would need more evidence to explain why the other result differs.’ The second sentence identifies both support and uncertainty.
Your vocabulary choice should fit the degree of evidence. ‘Shows’ describes an observation supported by data. ‘Suggests’ may suit a possible interpretation. ‘Proves’ is often too strong for a single small trial or an illustrative chart. The best word is the one your evidence can bear.
Keep time references intact
If a school programme improved its attendance last term, do not report that it ‘always increases attendance’. The evidence concerns one stated time period. A forecast for next term is also not an observation about next term. Words such as ‘was’, ‘is expected to’ and ‘has been approved’ do different work.
You can protect the timeline with a short clause: ‘In last term’s trial, participation was higher than in the previous term, but the next trial has not yet taken place.’ This is useful in English comprehension, formal writing and oral interaction, where a careless change of tense can alter a claim.
The earlier Guide 0408 on conflicting reports handles source attribution in writing. Here, the problem is different: how to preserve chronology while speaking clearly enough for the listener to follow it.
Explain one calculation without narrating every keystroke
When a numeric explanation is relevant, first state the relationship. ‘The percentage increase compares the change with the original amount.’ Then give the result. You seldom need to read out an entire formula and all intermediate digits in an oral response unless the question specifically asks how you calculated it.
A spoken mathematical explanation may be: ‘The price rose by twelve dollars from an original eighty. That is fifteen percent of the original price.’ The computation is transparent but the sentence remains natural.
If a question concerns a chemical demonstration, you might say: ‘The calculation uses the measured volume and the reacting mole ratio, so the final concentration depends on both the data and a correctly balanced equation.’ You do not need to reproduce an entire chemistry solution in English oral. You need to communicate the relevant point faithfully.
Deal with a follow-up that challenges the figures
An examiner may ask how your conclusion would change if a number were different. Do not simply repeat your prepared answer. Identify what changed, say how it affects the comparison, and keep the rest of your reasoning intact.
For example: ‘If only thirty of the fifty participants completed the survey, the completion rate would be sixty percent rather than eighty percent. I would still support offering the activity, but I would be more cautious about describing the trial as widely successful.’ This is a meaningful revision, not a dramatic retreat.
A prepared opinion is not a contract to defend every digit. It should be a position you can adjust when new information arrives. The ability to listen, reason and speak responsively is more useful than memorising one flawless paragraph.
Move from primary-school confidence to secondary-school precision
At primary level, a pupil may learn to give a complete answer with a reason. By upper secondary, a complex claim sometimes needs a reference group, a time frame or a limiting condition. The progression is not towards longer sentences for their own sake. It is towards more accurate explanations.
The PSLE English situational-writing route develops awareness of task, audience and source information. The Primary 6 English diagnosis route helps learners distinguish a weakness in language from one in reading the task. These are useful stepping stones for G3 oral speaking: decide what was asked, what information matters and how to make it intelligible.
This guide is not an invitation to speak with a rehearsed scientific accent. A calm voice, ordinary vocabulary and one clearly explained comparison are stronger than a string of technical terms that the speaker cannot unpack.
Prepare with a small evidence map
Use preparation time to write a few prompt words, not a manuscript. A simple map might be ‘view: useful if explained; evidence: two close readings, one outlier; meaning: checking matters; limit: not proof of accuracy; personal link: visitors learn to question results’. This map gives you a route while leaving you free to speak naturally.
If the stimulus does not show figures, do not invent them to fill the map. Replace numerical evidence with an observable detail or a carefully described hypothetical example. Say that your example is hypothetical if it needs that distinction.
Rehearse in two versions. The first version is longer and allows you to discover your strongest point. The second removes repetition and keeps only the explanation that the listener genuinely needs. The revised speech should sound more natural, not more densely memorised.
A four-week practice route
In week one, practise naming units and reference groups. Take simple figures from fictional worksheets, turn them into one complete spoken sentence, and check whether another person could explain what the number measures. Avoid statistics gathered without context.
In week two, practise spoken comparisons: counts versus rates, percentage changes versus percentage points, predicted outcomes versus observed outcomes. Give one example of each and add a sentence explaining the limit. Record yourself if it helps you hear where the sentence becomes crowded.
In week three, prepare brief oral responses to different kinds of video or community scenarios. Make a point, choose one useful detail, explain it and qualify it only when necessary. Practise a follow-up that changes a number or a condition. The aim is not to memorise your first answer but to adapt coherently.
In week four, practise with the official Paper 4 task types in mind: a response planned from a video stimulus and a two-way spoken interaction. Use an appropriate practice schedule set by your teacher, including a ten-minute preparation opportunity where relevant. Review your response for task fulfilment, clarity, natural delivery and accuracy of claims, rather than inventing a mark scheme.
A self-check a listener can use
After an answer, ask someone who did not see your notes to tell you what the main point was, which number mattered and what it was compared with. If they cannot, the problem may be the sequence rather than your vocabulary. If they misunderstand the units, insert the units next to the relevant quantity. If they remember the numbers but not your opinion, put your judgement earlier.
Keep a short record of errors. ‘Forgot the denominator’ is more useful than ‘poor fluency’. ‘Stated a forecast as a fact’ is more useful than ‘not confident’. ‘Read four values aloud when one comparison would do’ tells you exactly what to change in your next attempt.
A listener’s understanding is your final test. The answer should be sufficiently complete to stand on its own, but light enough to follow without a printed table or a calculator.
Practise fifteen short speaking problems
The cases that follow are written practice, not recordings from a real SEC examination. They use constructed numbers so that you can hear the difference between a figure, its meaning and a claim about what it proves. For each one, read the information, put it aside and give a short answer aloud. Then read the model reasoning and explain the risk behind the weaker version. You are not expected to memorise the sentences.
A completion rate needs its group
A fictional school event records 40 completed tasks among 50 participants. Explain this aloud to someone who has not seen the table.
Say: ‘Forty of the fifty participants completed the task, so the completion rate was eighty percent in this group.’ The group, task, count and rate are named together.
Do not say ‘eighty percent of students are successful’ because the record concerns a specific group and task. The phrase ‘this group’ prevents accidental generalisation from a limited sample.
Counts and proportions tell different stories
Group A has twenty completions among fifty people. Group B has six among ten. Compare both groups in speech.
Group A completed more tasks in total, twenty compared with six. Group B had a higher completion rate, sixty percent compared with forty percent. Both comparisons are useful for different questions.
A speaker who says ‘Group A performed better because twenty is larger than six’ ignores unequal group sizes. A speaker who says ‘B achieved more completions’ confuses proportion with total count.
One percent change, two directions
A price rises from $80 to $92. Someone calls it ‘a twelve-dollar or fifteen-percent rise’. Explain whether returning to the original price requires a fifteen-percent fall.
A clear answer says that fifteen percent is calculated using the original eighty dollars. Returning from ninety-two to eighty requires a reduction of twelve dollars, or roughly thirteen percent of ninety-two.
Do not tell the listener that the two percentages should match. The reference amount has changed. Keep the explanation short by naming the original and new bases.
Percentage points are not percentage increases
A fictional club’s attendance rate changes from forty to fifty percent. What wording preserves both possible comparisons?
Say: ‘Attendance rose by ten percentage points. Relative to the original forty-percent rate, that is a twenty-five-percent increase.’ One describes the gap between rates; the other describes relative change.
If the speaker says ‘attendance increased ten percent’, a careful listener may not know which measure was intended. Precision is part of comprehensibility, not decoration.
A precise concentration with plain language
A worksheet calculates the solution’s concentration as 0.1288 mol/dm³. The audience is not studying Chemistry.
You can say: ‘Under the stated reaction and volume measurements, the estimated concentration is about 0.129 moles per cubic decimetre.’ If necessary, add that this is an amount per litre of solution.
Avoid converting the result into an unsupported public safety statement. A concentration calculated in a classroom exercise does not by itself demonstrate that an unknown liquid is safe, drinkable or correctly identified.
A cluster and an outlier
A fictional demonstration gives readings of 20.10, 20.15 and 20.85 cm³. Explain the observation, not an invented cause.
Say: ‘The first two readings are close, only 0.05 cubic centimetres apart, while the third is substantially higher. That makes the selection and checking of results important.’
Do not assert that the third trial used faulty apparatus or that the first two are infallible. The table alone shows disagreement; it does not reveal the physical reason.
A recommendation without a guarantee
A fictional project estimates that one plan could use fewer materials, but the estimate excludes staffing costs. Give a responsible recommendation.
Say: ‘The first plan appears cheaper for the materials included in this estimate. I would consider it, but I would check staffing and safety requirements before calling it the lower-cost option overall.’
The speaker reaches a view while naming an important limit. Saying ‘the plan is proven best’ would exceed the evidence; refusing any decision would also be unhelpful.
A forecast is not a result
An organiser predicts 150 visitors for next month based on 120 visitors last month. Explain the statement carefully.
Say: ‘Last month there were 120 visitors. The organisers expect about 150 next month, but that figure is a forecast rather than an observed total.’
Avoid changing ‘expect’ into ‘achieved’. The grammar carries the evidence’s timeline. One word can transform a forecast into a false factual claim.
Explain a scale conversion
A display compares 72 kilometres per hour with 20 metres per second. How should the speaker relate the figures?
Say: ‘These speeds are equivalent. Seventy-two kilometres per hour is twenty metres per second, after converting both the distance and time units.’
Do not call one number bigger simply because seventy-two is numerically larger than twenty. The units differ. This case tests whether you explain meaning rather than compare bare digits.
The missing denominator in a claim
A student says ‘twenty more people attended this year’ without providing last year’s attendance or the total scale.
The absolute difference may be correct, but the audience cannot judge its relative size. Ask for the reference counts before claiming a large percentage improvement or a transformation in participation.
You can still say the attendance increased by twenty people if that is what was recorded. The correction is not to invent a missing denominator; it is to keep the claim at the level the evidence supports.
A public message about an unfinished trial
A school science exhibition is proposed for next term. A speaker describes it as a programme already evaluated by visitors.
Correct the timeline: ‘The school plans to run the exhibition next term. Its value could be assessed afterwards through visitor feedback, but no such results exist yet.’
This is not a reluctant answer. It suggests an appropriate way to obtain evidence while making clear that future feedback is not yet available.
A changed answer under a new condition
The examiner asks what you think of a display based on a ninety-percent approval rate, then says the figure came from only ten volunteers.
A strong reply adjusts the confidence: ‘The result is encouraging for those ten volunteers, but such a small group may not represent the wider school. I would still consider the display, while collecting more views.’
Do not repeat the original certainty. Also do not declare that the ten responses are useless. They show something about those respondents, not necessarily everyone else.
Selecting rather than reading ten measurements
A hypothetical result table has ten readings with similar values and one noticeably different result. The speaker is asked what makes a clear public explanation.
Choose the pattern: ‘Most measurements were close, while one differed. The display should show how the results were checked and why the final figure was selected.’ Only give a precise reading if needed to illustrate the gap.
The aim is to preserve the evidence while reducing listening load. Reading ten decimals is not the same as explaining an experiment.
A personal opinion supported without invented statistics
The question asks whether students should volunteer at science exhibitions. You have no relevant survey.
Say: ‘I think volunteering can be valuable because students practise explaining ideas and listening to unfamiliar questions. It should still be organised so that visitors receive accurate information.’
There is no need to manufacture a percentage of students who agree. A concrete, relevant reason is stronger than an unsupported global-sounding claim.
The end of a planned response
A student has already made one clear point, supplied an example and explained the limit. They want to finish decisively.
Conclude by returning to the question: ‘So I would support this kind of exhibition when it teaches visitors both the science and the value of checking evidence.’ This final sentence does not introduce another topic.
An ending should make the answer feel complete. Do not pile on a new unsupported statistic or apologise for the response’s length. ## A final practice conversation
Imagine that the examiner asks whether a fictional school should use a science demonstration to teach people about trustworthy information. You might answer: ‘I would support it, especially if students explain not just the result but how it was checked. One useful example would be showing that two measurements can agree closely while a third differs. Visitors could then see why scientists look at the full record instead of simply choosing a neat number. I would make the display accessible, because not every visitor will know the technical units, and I would avoid treating one demonstration as proof of a broad claim.’
Now imagine the follow-up: ‘Would it still be valuable if it had no numerical display?’ You could say: ‘Yes, because a clear explanation of the sequence and the evidence could still teach useful reasoning. I would use plain words and perhaps a question visitors could answer, but I would be careful not to pretend the demonstration measured something it did not.’ The second answer adapts to the new condition and keeps the same underlying principle.
This is the point of the guide. You are learning to think with evidence while speaking naturally, not to deliver a prewritten chemistry performance. English oral communication becomes clearer when each number has a job, each comparison has a reference and each conclusion has an appropriate boundary.
Frequently asked questions
Do I need to quote exact statistics in the K300 oral examination? No general rule says you must recite statistics. Respond to the actual stimulus and question. Use accurate numbers only when they are relevant and sufficiently supported.
Should I always describe every limitation? No. Name important qualifications that change the meaning of your claim. Too many minor disclaimers can hide the main answer.
Can I use everyday language instead of technical notation? Yes, when it communicates the meaning accurately. If a technical unit matters, name it and explain it simply rather than omitting it.
What if a follow-up changes the reference group? Recalculate or restate the comparison against the new group, then explain whether your judgement changes.
Is a prepared response the same as a memorised paragraph? No. Preparation should help you organise a relevant response to the actual video and prompt. Memorised examples can misfit the stimulus.
How does this connect to Mathematics and Science? Mathematics makes the comparison and precision sound; Science explains what was actually observed; English makes the reasoning clear to the audience.
What if I cannot remember a numerical detail? Avoid inventing it. Use the details you know, describe the relationship accurately or explain the principle without a fabricated statistic.
How can a parent or teacher help? Ask the learner to explain one result aloud, then repeat the main point back. If the listener cannot identify the quantity or reference group, the learner can revise the sentence instead of adding more facts.
The next step in the sequence
The next EMS volume brings the three subjects together in a fictional, text-only school demonstration audit. Chemistry will supply mole ratios, Mathematics will test units and rounding, and English will turn the checked findings into a visitor explanation. This is a deliberate progression from measurement and calculation to communication and integrated performance.
For official examination guidance, use the SEAB 2027 G3 syllabus list and the English K300 syllabus. For more on distinct earlier skills, see Guide 0408 on reconciling reports and Guide 0400 on written descriptions. Neither replaces the work of explaining numbers aloud.
The strongest spoken response is not the one with the largest number of figures. It is the one that allows a listener to understand a meaningful point, trust the reference and recognise what the evidence does and does not show.
