How can one free-fall demonstration support sound Physics, Mathematics and English thinking without allowing a convincing calculation to turn into an exaggerated claim? Begin with the raw quantities, check the mathematical model and then decide what readers may safely be told. This guide provides a complete original evidence-hearing case, a release standard, worked disputes, subject-transfer exercises and a revision pathway for 2027 G3 SEC learners.
Quick decision route: (1) label each measured quantity; (2) calculate speeds and acceleration with units; (3) test assumptions and measurement independence; (4) identify uncertainty and possible systematic bias; (5) distinguish the result from the conclusion; (6) publish only wording the evidence supports. Treat the figures below as constructed instructional data, not as actual test results.
Six evidence gates before the first public sentence
- Is each number measured, calculated or assumed?
- Does every length and time describe the correct event?
- Are the units and formulas appropriate?
- Does the model agree with the information without counting one premise twice?
- Do unresolved limitations affect the conclusion?
- Could another reader reproduce the reasoning and understand what is only illustrative?
What a useful evidence hearing does
A school group has measured the movement of a falling object and wants to announce that its experiment demonstrates gravitational acceleration. The physics calculation looks plausible; the mathematical graph looks neat; the student’s public statement sounds confident. None of those qualities alone settles whether the conclusion is defensible. This guide shows how to convene a simple evidence hearing: inspect the measurement, check the model, review the wording, name remaining uncertainty and decide what can be shared. That is the organising question throughout. You do not have to become an expert in three subjects at once. Work from the smallest factual claim toward the larger interpretation. A responsible release is not the absence of every limitation. It is a conclusion whose scope matches the evidence available and whose qualifications are useful rather than evasive. We use an original, idealised classroom case and then introduce realistic complications. It is a learning model, not an SEAB practical paper or a record of an actual investigation.
Start with the 2027 SEC assessment map
The Singapore-Cambridge Secondary Education Certificate begins examination use in 2027, with relevant subjects assessed at G1, G2 or G3 levels. The current 2027 G3 subject codes are Physics K323, Mathematics K310 and English Language K300. They are distinct assessed syllabuses: this EMS hearing is a study bridge, not a claim that SEAB administers a combined English–Mathematics–Science examination. Physics K323 has Papers 1, 2 and 3, weighted 30, 50 and 20 per cent respectively; its practical paper assesses experimental skills alongside conceptual knowledge. Mathematics K310 uses two written papers weighted 50 per cent each, with essential working required and a real-world context in the final Paper 2 question. English K300 assesses writing, comprehension, listening and oral communication through four papers. When a question asks about a Physics measurement, show scientific reasoning. When it asks for Mathematics, justify the calculation and model. When it asks for English comprehension or writing, distinguish a text’s evidence from its writer’s claim. The same careful mind operates in all three; the marking tasks remain different.
A one-page case file before any judgement
Our initial case has a compact sphere with an effective 0.020 m light-beam obstruction length. The upper gate registers a blocking duration of 0.010 s, while a lower gate registers 0.005 s. Their passage events are separated by 0.20 s. The distance between the relevant passage positions is 0.60 m. The lesson’s working assumptions are that the gates are aligned, the effective obstruction length is measured correctly, a suitable consistent timing convention is used and the test interval is short enough for the usual near-Earth free-fall approximation to be useful. None of the figures is a reported measurement from a real pupil’s lab. They were chosen because they allow a reader to see the meaning of each calculation without getting lost in arithmetic. The first stage is to construct an evidence register recording the raw numbers and their units. No claim should appear in the register that cannot be traced to either a direct observation or an explicitly labelled calculation.
Evidence register: three kinds of time
The most important clerical detail is that a light-gate investigation can contain different quantities all called time. A blocking duration is the interval during which a single gate’s beam is interrupted. A passage interval is the elapsed time between the object passing the two gates. A release-time measurement is the elapsed time from actual release to a later event. These intervals may be recorded by different sensors and may have different sources of delay. If someone places 0.005 s under the heading ‘time between gates’, the resulting acceleration can be wildly wrong even though the arithmetic is performed correctly. Use descriptive table headings: upper blocking duration / s; lower blocking duration / s; gate-to-gate passage interval / s; flag length / m. Then make a separate column for calculated speeds. In the hearing, ask not ‘Do the numbers look reasonable?’ but ‘What physical event does each column record?’ The question catches errors early.
Physics check: short-interval speeds
Using the effective obstruction length divided by each short blocking duration, the upper gate estimates speed at approximately 0.020/0.010 = 2.0 m/s downward, and the lower gate estimates 0.020/0.005 = 4.0 m/s downward. Because the object moves during the obstruction interval, these are short-interval average speeds used as estimates of local speeds, not exact instantaneous values. With an assumed 0.20 s between the relevant passages, the estimated change in downward velocity is 2.0 m/s and the acceleration estimate is 2.0/0.20 = 10 m/s² downward. The chosen value is consistent with the approximately 10 m/s² free-fall acceleration used in school-level near-Earth calculations. The evidence does not prove that the sphere experienced zero drag or that every timer was perfectly calibrated. A committee member should ask for raw records, definitions and a schematic before accepting a polished calculation. The scientific point is how the measurements relate to motion.
Mathematics check: mean speed and distance
From the separation of 0.60 m and the passage interval of 0.20 s, the average speed between the gates is 0.60/0.20 = 3.0 m/s. That number lies between the two estimated local speeds, 2.0 and 4.0 m/s, as expected for an ideal uniformly accelerated journey whose start and end speeds differ by that amount. Another consistency test is to calculate the expected interval distance as mean speed multiplied by time: 3.0 × 0.20 = 0.60 m. This circular-looking check only becomes independent if the underlying measurements or predictions are genuinely separate. If every quantity was derived from the same mistaken duration, agreement among formulas might conceal the shared mistake. Strong mathematical evaluation therefore records which values were actually measured and which were calculated from others. Numerical consistency is useful; it is not a substitute for independent verification.
Mathematics check: a whole-journey quadratic
Suppose the sphere starts from rest at a defined height, and we use the simplified model h = 5t², where h is downward displacement in metres and t is elapsed time since actual release in seconds. The model predicts h = 0.050 m at 0.10 s; 0.200 m at 0.20 s; 0.450 m at 0.30 s; and 0.800 m at 0.40 s. On a graph of h against t², the points form a straight line with gradient 5 m/s², giving a model acceleration of approximately 10 m/s². But this table is generated from the model, not independent experimental data, and therefore cannot be used as empirical confirmation of the model. If the laboratory obtains independently observed displacement–time pairs, those readings can be compared with predictions and their departures examined. A report that says ‘our ideal numbers prove our ideal equation’ mistakes a worked example for experimental evidence.
How to reconcile two approaches without double counting
The gate method estimates velocities at two positions and relates their difference to a measured time interval. The whole-journey method relates displacement to squared time under an assumption about the start. They can both be consistent with a constant-acceleration model. Their agreement is more persuasive when their measurements are independent and their calibration histories do not share the same hidden error. If both depend on a delayed release trigger or a wrongly measured flag length, the same incorrect premise may influence both conclusions. The hearing should draw a simple dependency map: which values came from a measuring instrument, which came from a physical dimension, which were calculated and which were assumed? This exercise may appear administrative, yet it is an advanced reasoning skill. It explains why two equations producing the same answer do not always count as two pieces of independent evidence.
The first red flag: a folded flag
In the original English story, a student notices that a sensor flag has a folded edge. That detail is not evidence that a particular speed was false; it is a reason to check the length used in its calculation. If the effective beam-obstruction distance were 0.025 m instead of the assumed 0.020 m during one run, a 0.010 s blockage would imply an approximate local speed of 2.5 m/s rather than 2.0 m/s. If both gates used the same wrong length, each speed would scale in the same direction. Whether the estimated acceleration changes depends on the actual geometry and intervals. The correction is not to choose whichever flag length produces the expected answer. It is to inspect the obstruction path and measure the relevant length with suitable equipment. A responsible hearing records the discrepancy and decides which previous numbers need recalculation.
The second red flag: a hidden time origin
An operator may begin a logger before the object is actually released. If the release occurs 0.030 s after the instrument’s nominal start, then a timestamp of 0.200 s after nominal start corresponds to about 0.170 s since release, assuming the clocks and trigger relationship are described correctly. Inserting 0.200 into a formula that assumes t = 0 at actual release will distort its prediction. By contrast, the electronically measured elapsed time between two later gate events can be less sensitive to the release offset, provided the same clock is used. These distinctions require careful causal reasoning, not elaborate notation. The panel should ask when the clock began, what event the logging system treats as a passage and whether the passage definition matches the measured gate separation. Correcting a time origin may matter more than adding extra decimal places.
The third red flag: extra digits and false precision
A screen can display several decimal places, but that display does not guarantee equal accuracy. Device resolution, alignment, calibration and definition of the measured event all affect the uncertainty. Suppose a 0.005 s blockage has a timing uncertainty of roughly 0.001 s. That hypothetical uncertainty is a substantial fraction of the measured duration; interpreting a speed estimate to many decimal places would be misleading. We cannot assign that uncertainty to the real apparatus without checking its specification. The broader principle is that short intervals magnify the importance of timing limitations. A sensible report retains sufficient working digits for calculations but rounds final findings to a precision supported by the inputs. It also separates a rounded numerical value from the confidence we have in the physical method. Printing ‘10.000000’ is not a substitute for a controlled experiment.
The fourth red flag: a picture without a clock
A photograph of the falling object may look dramatic and may be useful for readers, but one still frame does not demonstrate acceleration. Acceleration requires information about changes in velocity over time. A sequence of frames with known time intervals can support motion analysis; so can gate timings or an appropriate motion sensor. If a class poster presents a single photograph with the caption ‘Proof that the object accelerated at exactly 10 m/s²’, the caption goes beyond the evidence pictured. The correction is to state what the image actually records and link to the timed measurements separately. This is an English communication lesson as well as a scientific one: a reader should not be led to mistake illustration for measurement. Our public guide itself remains text-only, so the imagined photograph is a reasoning exercise rather than a media element.
The fifth red flag: the attractive straight line
A line through data points can look persuasive even if it has been fitted under an inappropriate assumption. On a graph of downward displacement against time squared, a straight line is expected for the selected rest-start constant-acceleration model. A non-zero intercept might suggest an offset in the displacement origin, but it can have other explanations. A steadily changing pattern of residuals can indicate a model mismatch or measurement bias. The panel should inspect axes, units, number and range of measurements, the method used to fit the line and the shape of residuals. A line fitted to four values generated directly from h = 5t² provides no new evidence. A line fitted to independently recorded trial data, with clearly described measurement uncertainty, is a more meaningful test. Never confuse the elegance of a graph with the quality of the underlying observation.
Evidence confidence: green, amber and red
A helpful classification does not claim universal scientific thresholds. Green means the particular claim is supported by identified measurements, calculations and appropriately stated limits. Amber means an important method question remains unresolved, so the proposed wording must be weakened or the claim withheld pending a check. Red means the claim contradicts known evidence or depends on a demonstrably false measurement assumption. These colours are a classroom decision aid, not an official SEAB marking scheme. For instance, ‘The two gate readings are consistent with increasing downward speed in the illustrative model’ is green relative to the example’s stated assumptions. ‘We proved air resistance did not act’ is red: the supplied observations do not establish that. ‘Our real apparatus produced exactly these idealised results’ is also red if the numbers are fabricated teaching values rather than recorded data. The panel should justify each classification with evidence.
Writing a public conclusion without overclaiming
The clearest defensible public statement may be short: ‘Using a known obstruction length and electronically timed light-gate intervals, our idealised example gives approximate downward speeds of 2.0 m/s and 4.0 m/s and an acceleration estimate of 10 m/s². This is consistent with the school-level constant-acceleration model; actual measurements require checks of flag length, timing and alignment.’ The distinction between example and observation belongs in the statement itself, not hidden in a footnote. A weaker conclusion would say the class has proved an exact law with a flawless device. An evasive conclusion would say nothing can ever be known. Responsible language is precise and proportionate. The English writer must help the reader understand what the numbers mean without transforming caution into a wall of qualifications. Each limitation should answer a real question about the evidence.
A release decision with conditions
Imagine the group wants to post a one-page notice to classmates. The review committee can approve a notice about the method and the worked example immediately, because its numbers and assumptions can be checked in the text. It should not approve the sentence ‘Our live trial confirms the exact free-fall acceleration’ without access to the raw experimental record and apparatus checks. A conditional decision might request a measured flag length, a labelled timing trace, at least several repeat trials and a reworded conclusion. The difference between a publication decision and a measurement decision matters. Students can legitimately teach the method using constructed data if they say those data are constructed. They cannot honestly present constructed values as observations. The panel’s job is to make a useful explanation available while preserving the boundary between learning illustration and evidence from the world.
Exam transfer: Physics, Mathematics and English
In a G3 Physics response, a teacher or examiner may ask for calculation, explanation, a fair-test suggestion, the significance of a graph or an evaluation of the experimental method. The answer should follow scientific conventions: quantities, units, cause, evidence and limitation. In G3 Mathematics, the focus is on variables, relationships, algebra, graph labels, suitable precision, reasonableness and essential working. In G3 English, a candidate may need to follow a narrative’s sequence, recognise an inference, explain language effect or write accurately for an audience; a scientific setting is only one possible context. The official examination requirements are subject-specific. The integrated hearing strengthens transfer because it makes students ask a common question: ‘Which evidence makes this answer justified?’ But it does not replace subject practice, authentic syllabus consultation or teacher feedback.
The habit that lasts beyond the examination
A disciplined learner can become more confident without becoming more certain than the evidence permits. The basic routine is to read the problem carefully, label the information, calculate transparently, test the result against the context and make a conclusion no larger than the information supports. When feedback shows an error, revise the reasoning step instead of merely memorising the corrected number. In later subjects and work, similar habits help readers assess a chart in a news article, compare technical proposals and question a plausible-sounding claim. This is why a small free-fall case is a useful teaching vehicle. It gives a concrete reason to care about precise mathematics, accurate science and plain English. In a mature answer, clarity is not decoration added after the thinking. It is evidence that the thinking has been organised enough to be examined.
Evidence-hearing docket: twenty-two decisions
These original mini-cases are deliberately different. Some have arithmetic mistakes; others contain measurement, model, narrative or public-language problems. For each one, decide what can be accepted, what should be corrected and what extra evidence is needed. The explanation follows the case. The objective is not to mistrust every observation but to make each conclusion proportionate.
1. Wrong denominator in the upper speed
The flag length is 0.020 m and upper blocking duration is 0.010 s. A note says 0.020 divided by 0.20 gives the upper-gate speed. The calculation uses the between-gates travel interval rather than the local blockage interval. That division gives 0.10 m/s, a number unrelated to the requested local estimate. The appropriate speed estimate is 2.0 m/s. The panel should flag the denominator, correct the table heading and ask the student to restate the meaning of 0.20 s. Recomputing is necessary but insufficient unless the confusion is repaired.
2. Inconsistent metres and millimetres
A team records the flag length as 20 mm but enters the numeral 20 into a formula expecting metres. It obtains a spectacularly large speed estimate. Before considering new physical explanations, convert 20 mm to 0.020 m and recalculate. Correct units often identify an error faster than a complicated investigation of forces. A defensible report shows the converted value with the unit at the substitution step so that another reader can audit it. A number without units should never be released as a measured speed.
3. Two speeds, one acceleration
The upper and lower approximate downward speeds are 2.0 m/s and 4.0 m/s, with a passage-time difference of 0.20 s. The estimated acceleration is (4.0 − 2.0)/0.20 = 10 m/s² downward. This is a model-consistent result. It does not establish that the apparatus is calibrated or that there was no resistance. The final sentence should distinguish the calculated magnitude from the experimental certainty. The hearing marks this calculation as arithmetically consistent under the stated assumptions.
4. Mean speed is not final speed
The gates are 0.60 m apart and the travel interval is 0.20 s. The between-gates mean speed is 3.0 m/s. A display label calls this the ‘lower gate speed’. That label is incorrect because 3.0 describes the whole interval, whereas the lower gate’s short blocking measurement estimates 4.0 m/s. Change the label before publication. A one-word heading error can make two valid calculations appear to conflict.
5. Quadratic line with an invented dataset
A slide contains four points exactly matching h = 5t² and calls them ‘observed acceleration data’. The points were produced by substituting time values into the equation, not by timing a real object. The line may illustrate the model but cannot independently validate it. Replace ‘observed’ with ‘idealised calculated’. If actual measured readings become available, make a separate table and discuss how closely they follow the model. This is a critical boundary between teaching mathematics and reporting science.
6. The pretty graph with the wrong axis
A graph uses values 0.0100, 0.0400, 0.0900 and 0.1600 on the horizontal axis, but the label reads ‘time / s’. Those values are times squared in s². Correct the axis. Otherwise a reader may interpret the gradient as a rate of displacement per second rather than per second squared. The panel should never approve a gradient-based claim until both axes and their units are checked.
7. A shifted release clock
A logger starts 0.030 s before the object is actually released. Someone reports t = 0.200 s since release simply because a timestamp reads 0.200 s. If the log and release signal refer to the same clock, the time since release is 0.170 s, not 0.200 s. The case asks for careful definition of a time origin; it does not require a complex algebraic model. A correct release note states whether timing began at the button press, actual mechanical release or first gate event.
8. Folded flag, plausible result
The flag’s nominal length is 0.020 m but it may have bent so the effective obstruction path differs. The resulting calculated speed happens to be close to a textbook value. Closeness to expectation does not fix the flag. The panel should request direct inspection or a controlled replacement flag and repeat the trial. Avoid choosing an adjusted effective length solely to make the answer equal 10 m/s²; that would reverse the logic of measurement.
9. Three inconsistent repeats
Three acceleration estimates from a hypothetical apparatus are 9.8, 10.1 and 11.7 m/s². The third is noticeably different, but the reviewer should not automatically discard it because it is inconvenient. Inspect its raw time measurements, alignment and possible release disturbance. Calculate summary statistics if requested, and record any exclusion rule with a reason. Retest where appropriate. An outlier can reflect a genuine unusual event or a data-handling problem; the observation alone cannot tell us which.
10. Systematic error survives averaging
Five trials all used a flag incorrectly recorded as 0.020 m when its actual effective length was 0.025 m. The speeds calculated from every trial share a length-related bias. More repeats can make the arithmetic average more stable while leaving it systematically wrong. The hearing should request calibration or direct remeasurement. This case tests why repeatability and accuracy are not interchangeable. A statement that ‘we did five trials, therefore the speed is correct’ gives repetition more authority than it deserves.
11. Residual pattern across time
Independent measured displacements at successively later times fall increasingly below the ideal predictions of h = 5t². Possible causes include model simplification, a timing offset or changing effects of air resistance, among others. Do not announce a single cause from the pattern alone. The panel can approve the descriptive statement ‘negative residuals increased over the observed range’ and request additional checks before endorsing the mechanism. The words ‘consistent with’ help keep possible explanations separate from established observations.
12. A photograph sold as proof
A school poster shows one still image of a sphere above the lower gate, with the headline ‘This picture proves acceleration.’ The image establishes position at an instant, not a change of speed over time. The correct improvement is to pair an accurately labelled image with independent time information, or remove the unsupported proof claim. The panel approves a caption describing the visible scene, not an inference the image cannot support. This example is about evidence and wording, not graphic design.
13. An extra zero after a rounded reading
The reporter changes ‘approximately 10 m/s²’ to ‘10.0000 m/s²’ because it looks more scientific. Those extra zeroes imply greater precision without adding measurement information. Restore the precision justified by the input quantities and device. If the numbers are idealised, say so. The committee should not confuse a calculator’s ability to print digits with an experiment’s ability to determine them accurately.
14. Evidence that supports a limited claim
A full raw record includes gate signals, measured obstruction length, the timing convention, repeated trials and a consistent calculation. The values broadly agree with a simple free-fall model. The statement ‘The observations are consistent with increasing speed in the downward direction over this interval’ is defensible. The stronger statement ‘All falling objects must have exactly these measured speeds’ is not. The first claim fits the case, while the second crosses into an unsupported generalisation. Good editorial judgement is deciding how large a claim the data can carry.
15. Narrative confidence versus author judgement
An original narrative presents a student saying ‘The numbers will speak for themselves’ before discovering mislabeled columns. In an English comprehension response, the sentence reveals the character’s premature confidence. It should not be treated as the author’s scientific assertion. In a public report, however, the same words might mislead a reader if presented without the later correction. Context changes how we interpret language; evidence remains the standard for deciding what a factual report may claim.
16. A plausible statement with no source
The proposed notice says ‘The official SEC examination requires a light-gate practical on falling spheres.’ No such required named practical follows from the official syllabus information provided here. Specific examination questions can differ. The panel removes the assertion and replaces it with a safer, useful statement that the Physics syllabus assesses relevant motion concepts and experimental skills. Students should prepare transferable skills rather than claim knowledge of an unseen paper.
17. A rushed public announcement
A team has completed calculations but has not verified the effective flag length or timing convention. It wants to publish ‘We experimentally proved gravity is exactly 10 m/s².’ The panel should withhold that claim. It may publish a labelled worked example explaining how speed and acceleration are estimated, while noting that apparatus checks are still needed for a measured conclusion. This is not an instruction to hide work. It is a decision to present what is genuinely ready.
18. Independence makes agreement meaningful
A teacher verifies the flag length using a separate measurement and a second logger independently measures the between-gates interval. Results agree with the original measurements within known limits. That agreement is more informative than calculating the same acceleration twice from an unchanged spreadsheet. The committee should document what was independent and why it matters. Do not call different formulas independent evidence if they draw entirely on the same raw values and assumptions.
19. One technical word changes the claim
Compare ‘The data proved acceleration’ with ‘The data were consistent with approximately constant downward acceleration over the measured interval.’ The latter gives the reader the quantity, direction, scope and degree of confidence. It is longer because it communicates necessary conditions, not because it is trying to sound sophisticated. The panel should favour this kind of plain precision over empty certainty and over a paragraph of legalistic caution.
20. Which answer deserves a mark?
A student writes only ’10’. The arithmetic might match the expected magnitude, but the response does not show formula, unit, direction, working or relation to the question. Another student writes ‘(4.0 − 2.0)/0.20 = 10 m/s² downward’ and adds a short caveat where evaluation is requested. The second is easier to assess and more faithful to both the calculation and the real-world story. Examinations reward the requested reasoning, not the ability to print a plausible number.
21. A last-minute corrected table
The report’s table originally uses the heading ‘travel time’ for both blockage duration and gate-to-gate interval. Someone notices this just before release. Revising the headings is not a superficial copyedit; it changes how readers interpret the entire calculation. Recheck each ratio, unit and result in the updated table rather than assuming that a label change has no mathematical consequences. A sound correction process follows the error to every dependent claim.
22. A decision after an inconclusive check
The class repeats the trial but obtains variable readings because the sensor alignment keeps changing. The team should not fill gaps with ideal numbers and claim the problem has been solved. It can instead describe what was attempted, why the results are uncertain, and which adjustment is needed before a firm conclusion. In examination writing, this is a stronger evaluation than inventing a successful experiment. Honest uncertainty can still teach readers a great deal about the method.
A full integrated case to practise independently
The student council wants a public notice describing an idealised demonstration. The draft note states that a 20 mm flag interrupted an upper light gate for 0.010 s and a lower light gate for 0.005 s. The passage interval between gates was 0.20 s and their separation was 0.60 m. The group also supplies a displacement model h = 5t², with metres and seconds, and a fictional narrative in which a student realises that two columns labelled ‘time’ describe different events. As review chair, calculate both local speed estimates, the between-gates mean speed and the acceleration estimate; distinguish the model from independent evidence; identify the narrative’s important inference; write a short public statement and request one targeted verification step.
Worked response, numerical part: with the 0.020 m effective obstruction length, the two approximate downward local speeds are 2.0 m/s and 4.0 m/s. The between-gates mean is 0.60/0.20 = 3.0 m/s; the estimated acceleration is (4.0 − 2.0)/0.20 = 10 m/s² downward. The simplified model predicts h = 0.45 m at 0.30 s after actual release, provided the chosen time origin and rest-start assumptions hold. That prediction is not another observed reading. The narrative inference is that the character moves from premature confidence toward careful documentation when she understands her ambiguous headings. An appropriate next verification is to check the flag’s effective obstruction length and to compare it with the value entered in each speed calculation.
A concise notice could say: ‘This worked example shows how electronically timed intervals and a known obstruction length can give speed and acceleration estimates consistent with the near-Earth free-fall model. The values here are constructed to teach the method. A real measurement would need clear event timings, reliable length measurements and repeated checks before supporting a stronger experimental claim.’ The release decision is therefore approve as a labelled learning example, withhold the claim of proven experimental precision, and invite a later verified practical report if real data are collected. Notice the decision distinguishes communication readiness from experimental completeness.
A ten-day progression from foundations to independent judgement
Session 1 — Day 1: sort the evidence
Draw a two-gate setup and create five different fields for length, two blockage intervals, between-gates time and separation. Do not calculate yet.
Session 2 — Day 2: calculate local motion
Convert to metres and seconds, estimate local speeds and acceleration, then explain why each estimate is approximate.
Session 3 — Day 3: compare average and local speed
Calculate interval mean speed, identify the time window and describe why three sensible speeds can be different.
Session 4 — Day 4: build the quadratic model
Construct a displacement–time-squared table, label axes and recognise that model-generated points cannot verify the model.
Session 5 — Day 5: inspect uncertainty
Select one flag-length error and one trigger-time error. Say which calculations each would affect and propose a targeted check.
Session 6 — Day 6: test a claim
Write three conclusions: clearly false, overly confident and well qualified. Explain the differences in one paragraph.
Session 7 — Day 7: hold a small hearing
Give a partner the raw case and ask them to review your quantities, working and wording. Record the first unresolved point.
Session 8 — Day 8: repair an error
Deliberately choose a flawed calculation from the worked stations and trace the consequences into the final public conclusion.
Session 9 — Day 9: answer by subject
Write a concise Physics evaluation, a Mathematics model solution and an English explanation of a character’s change in confidence.
Session 10 — Day 10: independent transfer
Try an unfamiliar timed investigation. Use the same method without copying the specific free-fall numbers.
Frequently asked questions
Are EMS evidence hearings an official SEC paper?
No. This is a connected learning exercise. English K300, Mathematics K310 and Physics K323 have their own assessments and expectations.
What is the headline numerical result?
In the idealised example, short-interval speed estimates are 2.0 and 4.0 m/s downward and the acceleration estimate is 10 m/s² downward.
Does matching 10 m/s² prove zero air resistance?
No. A matching approximate result is consistent with the simple model under the stated assumptions, not proof that all other influences are absent.
Is h = 5t² a directly observed fact?
Not in this worked example. It is a chosen idealised model. Independent measurements are needed to test the model experimentally.
Why distinguish measured and calculated values?
A calculated quantity inherits assumptions and possible errors from its inputs. Distinguishing sources makes claims easier to audit.
Can we publish a worked example before testing apparatus?
Yes, if it is explicitly labelled as a teaching illustration rather than a record of actual measurements.
Does a good conclusion need many caveats?
No. It needs the specific limitations that materially change the reader’s understanding of the claim.
Why does this matter in English?
English examination and public communication both reward accurate comprehension, appropriate inference and clear expression, though the tasks differ.
What comes after this hearing?
Return to the weakest owner guide: Physics Volume 0930, Mathematics Volume 0931 or English Volume 0932. Practise the exact reasoning you missed.
What should a parent ask a learner?
Ask what was measured, what was assumed, which calculation produced the conclusion, and what one additional check would improve confidence.
Continue through the subject owners
Read Volume 0930: Physics light-gate measurements for local speed, acceleration and physical error; Volume 0931: Mathematics quadratic motion tables for graph models, interval speed, residuals and bounds; and Volume 0932: English narrative inference for sequence, character attitude and language effect. The preceding Volume 0929 pendulum evidence review is a useful comparison: the principles of fair testing persist, while the measured quantities and models change.
To rebuild earlier foundations, use Secondary 1 Science measurement records, Secondary 1 Mathematics foundations and Secondary 1 English reasoning. Families exploring how the learning route starts can consult Sengkang PSLE results and Secondary 1 posting. For broader G3 choices, see Sengkang’s 2027 G3 English, Mathematics and Science exam guide.
Official 2027 syllabus references
Examination details and codes follow SEAB’s 2027 G3 school-candidate syllabus list (last updated 26 October 2026), the K323 Physics syllabus, the K310 Mathematics syllabus and the K300 English Language syllabus. This EMS hearing is an original cross-subject learning resource, not an official component of the SEC examination. Consult SEAB and your school for the current requirements of the exact subjects for which you are registered.
