English:GCSE science mathematics, graphs and uncertainty - Science Year 10
Introduction

Convert science units, choose and interpret graphs, calculate means and gradients, and explain uncertainty without confusing repeatability with accuracy. Year 10 placement is editorial within the GCSE pathway.
Convert units and report meaningful digits
A number needs a quantity, unit and appropriate reporting precision. Write a table heading such as time in seconds rather than leaving readers to guess whether 40 means seconds, minutes or a count. Quantities with different units cannot be added as though they were the same kind of amount. A calculator display does not determine how many digits are justified by the measurements.
Prefixes change numerical scale without changing the physical quantity. Kilo represents a thousand, milli a thousandth and micro a millionth of the unprefixed unit. Symbols are case-sensitive: M for mega and m for milli are not interchangeable. A millimetre is one thousandth of a metre, but square and cubic conversions must apply the factor to every dimension.
One centimetre is one hundredth of a metre. Squaring gives one square centimetre as one ten-thousandth of a square metre; cubing gives one cubic centimetre as one millionth of a cubic metre. Consequently 2.40 square centimetres is 0.000240 square metres, while 3.20 grams per cubic centimetre is 3200 kilograms per cubic metre, best written 3.20 times ten cubed to preserve the stated three significant figures. Convert both numerator and denominator of a compound unit.
Use standard form with a coefficient from one up to, but not including, ten, multiplied by an integer power of ten. The number 0.00450 kg is 4.50 times ten to the minus three kg, or 4.50 g. It has three significant figures: leading zeros locate the decimal point, while the final zero here communicates reporting precision. An integer such as 12000 can be ambiguous; 1.20 times ten to the four states three significant figures explicitly.
For an original supplied example, a distance of 2.40 times ten squared millimetres is 0.240 m and a time of 0.800 s gives average speed 0.300 m/s. The unconverted ratio 240 divided by 0.800 gives 300 millimetres per second, which is the same physical speed when converted. It is not 300 metres per second. Keep extra digits during intermediate calculations, then apply the appropriate final rounding; do not round every intermediate result to one digit.

Resolution concerns the smallest distinguishable change in a reading under the stated method. Accuracy concerns closeness to the relevant quantity value; repeatability concerns agreement under the same conditions. More decimal places do not automatically mean better accuracy. A finely divided scale can still have a zero offset, and a digital display can show stable but biased results.
Record the raw values and their units before calculating. Exact counts, such as the stipulated twenty complete oscillations, differ from measured quantities; that exact count does not itself restrict the significant figures in a mean period. A result should identify what was measured, how it was obtained and what its uncertainty statement means, rather than hiding important information behind a long decimal.
Choose graphs and interpret gradients, intercepts and rates
Choose a graph for the question and the type of data. A bar chart compares categories; a scatter graph examines association between paired quantitative variables. A continuous-variable experimental graph can show measured points and an appropriate best-fit line or curve. Joining every noisy point mechanically is not the same as fitting a justified relationship. State axis quantities and units, use a readable scale and do not silently omit inconvenient values.

For this original fictional spring comparison, put applied force F on the horizontal axis in newtons and extension x on the vertical axis in millimetres. The supplied records are exact model values for learning the graph arithmetic, not a measured spring test. Assume the illustrated comparison is within the elastic range. Column B has a constant reading offset; it is not an extra applied force.
| Applied force F in N | Model A extension x in mm | Model B indicated extension in mm |
|---|---|---|
| 0.00 | 0.00 | 0.40 |
| 1.00 | 0.80 | 1.20 |
| 2.00 | 1.60 | 2.00 |
| 3.00 | 2.40 | 2.80 |
| 4.00 | 3.20 | 3.60 |
Model A is a straight line through the origin, with extension increasing by 0.80 mm for each newton. Model B is equally straight but starts at 0.40 mm and follows the same increase per newton. A straight graph is not enough to establish direct proportion: the relevant line must also pass through the origin within the evidence and uncertainties. A zero offset is one possible explanation for an intercept, not a universal diagnosis for every graph that misses the origin.
Here g is graph gradient and x and y denote its horizontal and vertical coordinates. In this particular plot, horizontal coordinate is force and vertical coordinate is extension; the letter x used for physical extension above must not make you swap the axes. Using model-A points at 0 N, 0 mm and 4 N, 3.20 mm gives gradient 0.80 mm/N, or 0.000800 m/N. That is compliance in this ideal model. The spring constant for force divided by extension is its reciprocal, 1250 N/m, not 0.80 N/m.
For real noisy data, choose widely separated suitable points on the fitted line, not necessarily two measured points, to reduce reading sensitivity. Label the gradient units as vertical-unit divided by horizontal-unit. A nonzero intercept generally makes a single ratio y divided by x different from the gradient. Do not force a best-fit line through the origin solely because that makes the calculation easier; explain whether the physical model and data justify it.
A curve has a changing gradient, so an overall chord gives an average rate between two stated inputs, while a tangent estimates a local rate. In a distance-time graph the gradient has speed units; in a cumulative reaction-gas-volume versus time graph it has volume-per-time units, not automatically concentration-per-time. The same slope calculation can represent different physical quantities depending on the axes.
Correlation does not by itself prove causation. Controlled conditions, plausible mechanisms, repeat evidence and alternative explanations matter. A graph showing larger values for two variables could reflect a third factor or a biased sample. Reading beyond the measured range is extrapolation and is less secure than describing the observed interval; no straight segment guarantees the same law will continue indefinitely.
Evaluate repeat data, corrections and uncertainty limits
Repeats help reveal variation, but they do not remove every error. Consider this original fictional record of time for twenty complete oscillations under the same conditions. The count of twenty is exact for the model. The values do not describe the stopwatch photograph or measurements taken from either video.
| Repeat | Complete oscillations counted | Elapsed time in s |
|---|---|---|
| 1 | 20 | 39.6 |
| 2 | 20 | 40.0 |
| 3 | 20 | 40.4 |
The mean elapsed time is 40.0 s, and division by twenty gives a mean period of 2.00 s. The range is 40.4 minus 39.6, or 0.8 s. Under the explicitly chosen classroom half-range convention, the repeat-spread estimate is 0.4 s for the elapsed time, or 0.020 s for the period; relative to the mean each is one percent. This is not a complete instrument/reaction-time uncertainty budget, a standard deviation, a standard error or an automatically justified ninety-five-percent confidence interval.

Timing several oscillations can reduce the relative contribution of a fixed start/stop timing contribution when divided by the count, but longer runs may introduce counting mistakes or changing conditions. Averaging repeated values can reduce the influence of random variation in the mean under suitable assumptions. It does not automatically remove a common zero offset or other systematic effect. Random error describes unpredictable variation; systematic effects can shift many readings in a common way.
Suppose a separate supplied calibration check shows this timing method adds a fixed 0.60 s to every elapsed-time value. Subtracting that known correction from the original mean gives 39.4 s, so the corrected mean period would be 1.97 s. The raw repeat spread is unchanged by subtracting the same constant. Real calibration values have uncertainty; do not treat this fictional stipulated correction as proof that every real bias is known exactly.
Preserve an unexpected reading. If an additional recorded time were 44.0 s and all four values were initially treated as twenty-oscillation runs, their mean would be 41.0 s and the reported period 2.05 s. That change is a reason to inspect the method and record, not to delete 44.0 merely for spoiling a neat pattern. If the original observer’s note proves it counted twenty-two oscillations, document that mismatch; it cannot be pooled as an equivalent twenty-count run. Do not invent such a note simply to justify rejection.
For a separate original speed-bound exercise, stipulate distance 0.800 m with limits 0.790 to 0.810 m, and elapsed time 0.400 s with limits 0.380 to 0.420 s. The central ratio is 2.00 m/s. Because both quantities are positive, the smallest possible ratio under the stated independent allowed bounds is 0.790 divided by 0.420, about 1.881 m/s; the largest is 0.810 divided by 0.380, about 2.132 m/s. An outward-rounded enclosing interval is 1.88 to 2.14 m/s. These are supplied hard model bounds, not probabilistic confidence limits, and their ratio is not found by dividing the two uncertainty widths alone.
Observe the spring-mass oscillation context and distinguish the variable being changed from the period being timed. The video includes more advanced formulae; deriving them is not required here. Compare with the course’s separate fictional twenty-oscillation table rather than pretending its numbers were measured from the video. Original: Flipping Physics, Demonstrating What Changes the Period of Simple Harmonic Motion(SHM). The written explanations and native tasks work without sound or the external player.
Review the evidence cycle: Science Buddies — The Scientific Method: Steps and Examples. Relate questions, data, interpretation and further checks. The original repeat table and worked feedback provide the offline alternative; watching a method does not mean you have carried out a practical investigation.
State a conclusion at the strength the evidence supports. Report the chosen calculation, meaningful digits, units, range and uncertainty convention; name systematic effects or omitted contributions that averaging cannot fix. A graph, a small spread and a correct-looking decimal are useful evidence only when the method and comparison boundary are also clear.
Interactive Tasks
Quiz: Test Your Knowledge
How many cubic metres equal one cubic centimetre? (One millionth of a cubic metre) (!One hundredth of a cubic metre) (!One hundred cubic metres) (!One thousandth of a cubic metre because cubing changes nothing)
What is 0.00450 kg in grams with its stated reporting precision? (4.50 g with three significant figures) (!450 g with no need to change the unit factor) (!0.00450 g because only the label changes) (!4.5 g with a required loss of the final significant zero)
What is the original 0.240 m divided by 0.800 s speed? (0.300 m/s) (!300 m/s because millimetres are metres) (!3.00 s/m as the same quantity) (!An accuracy guarantee from the calculator digits)
Which graph tests association between two quantitative variables? (A scatter graph with labelled quantities and units) (!A bar chart whose bars are unlabelled categories) (!A picture that does not encode paired values) (!Any curve drawn without the observations)
What is model A gradient on extension in mm versus force in N axes? (0.80 mm per N, equivalent to 0.000800 m per N) (!0.80 N per m regardless of axis units) (!1250 mm per N because the reciprocal has the same units) (!A dimensionless number because both axes contain numbers)
What distinguishes the original model B from direct proportion? (Its straight line has a nonzero 0.40 mm intercept) (!It has the same gradient so it must pass through the origin) (!Its existence proves every physical intercept is a zero error) (!It has no possible graph because all straight lines pass through zero)
What is the original mean period from the three twenty-count runs? (2.00 s from mean elapsed time 40.0 s divided by twenty) (!40.0 s because the count is irrelevant) (!0.800 s because the range equals the period) (!A guaranteed true period known with zero uncertainty)
What does the chosen half-range calculation provide here? (A stated repeat-spread estimate rather than a complete uncertainty budget) (!An automatic ninety-five-percent confidence interval in every investigation) (!Proof that all systematic effects are absent) (!The same mathematical quantity as a standard deviation for any data)
What happens to the mean period under the separate stipulated 0.60 s timing correction? (It becomes 1.97 s while the raw repeat spread is unchanged) (!It becomes 2.60 s because the correction is added to each period) (!It stays 2.00 s because averaging eliminates every bias) (!It certifies the actual photographed stopwatch is perfectly calibrated)
How are the supplied positive speed limits calculated? (Lowest distance over highest time, and highest distance over lowest time) (!Divide the distance uncertainty width by the time uncertainty width only) (!Use the central ratio as an exact value for all allowed inputs) (!Delete the lower and upper bounds when a calculator shows three digits)
Memory Game
| Resolution | Smallest distinguishable reading change under the stated method |
| Repeatability | Agreement of repeated measurements under the same conditions |
| Accuracy | Closeness to the relevant quantity value |
| Gradient | Change in vertical coordinate divided by change in horizontal coordinate |
| Systematic effect | A common-direction influence not automatically removed by averaging |
| Extrapolation | Extending an inferred relationship beyond the observed range |
Drag and Drop
| A cubic centimetre is converted to cubic metres | Cube the centimetre-to-metre factor |
| A graph uses extension vertically and force horizontally | Report gradient in extension units per force unit |
| A fitted straight line has a nonzero intercept | Do not claim direct proportion solely from straightness |
| Repeated timing values cluster tightly | Still investigate common bias and omitted uncertainty contributions |
| An unexpected timing changes the mean | Retain the raw value and investigate before any justified exclusion |
Crossword Puzzle
| GRADIENT | What describes rise divided by run on specified graph axes? |
| PREFIX | What attaches a decimal factor to a unit symbol? |
| MEAN | What is the sum of repeated values divided by their count? |
| RANGE | What is maximum minus minimum in the repeat record? |
| INTERCEPT | What is the vertical-axis value where the horizontal coordinate is zero? |
| UNCERTAINTY | What describes limitations in knowledge of the measured quantity rather than a known mistake? |
LearningApps
Optional extra practice. Choose an appropriate activity with your teacher. If the external collection is unavailable, use the native activities and independent tasks in this course.
Cloze Text
Open Tasks
Easy
- Convert prefixes: Explain kilo, milli and micro and convert the original mass into grams without discarding reporting precision.
- Read the illustrations: Distinguish the two ruler scales, stopwatch dials and an unlabelled coordinate grid from actual supplied measurements.
- Define measurement quality: Distinguish resolution, repeatability and accuracy using a plausible stable zero-offset example.
- Choose a graph: Select appropriate displays for categories and paired quantitative values with axis units.
Standard
- Convert areas, volumes and density: Convert the supplied 2.40 square centimetres and 3.20 grams per cubic centimetre with complete factors.
- Calculate and interpret gradients: Use both original model-A endpoints, convert units and distinguish compliance from reciprocal spring constant.
- Analyse repeat timing: Calculate mean, period, range and the explicitly chosen half-range estimate without calling it a complete budget.
- Apply the stipulated correction: Calculate the corrected elapsed mean and period, explaining why subtracting a fixed amount does not change spread.
Advanced
- Interpret an intercept: Compare the two original graphs and discuss whether offset, background or a different model needs further evidence.
- Audit an unexpected value: Recalculate with the extra 44.0 s, retain all raw data and explain when an observed count mismatch could justify separate handling.
- Bound a derived quantity: Calculate the extreme speed ratios from the supplied hard limits and round the interval outward.
- Evaluate a scientific claim: Identify why correlation, a small repeat range and many calculator digits do not by themselves establish causation or high accuracy.
Learning Check
- Units and meaningful digits: Apply factors to every dimension and preserve stated precision.
- Original speed and graph axes: Retain physical quantities and units through calculation.
- Intercept and changing rates: Distinguish linearity, proportion and local gradient.
- Repeated periods and correction: State both arithmetic and the limited meaning of spread.
- Unexpected data: Preserve the audit trail rather than deleting an inconvenient observation.
- Bounds and justified conclusions: Use positive ratio extremes and outward rounding.
Worked checks and assessment guidance
- One cubic centimetre is ten to the minus six cubic metres, while 2.40 square centimetres is 0.000240 square metres. The supplied 3.20 grams per cubic centimetre becomes 3.20 times ten cubed kilograms per cubic metre. The mass 0.00450 kg is 4.50 g with three significant figures. Leading zeros do not count as significant digits, but the final stated zero in 4.50 does. A calculator cannot establish measurement quality.
- The supplied 240 mm is 0.240 m and 0.240 divided by 0.800 gives 0.300 m/s. A scatter graph can examine paired quantitative data. The original extension-versus-force slope is 3.20 mm over 4 N, 0.80 mm/N, equivalent to 0.000800 m/N. Its reciprocal is the ideal spring constant 1250 N/m. The generic Commons coordinate grid is not a measured plot of this table.
- Model A passes through zero; model B has the same 0.80 mm/N gradient but a 0.40 mm intercept, so it is not directly proportional as indicated. One possible reading offset is supplied for this fictional comparison, but other real intercepts require contextual evidence. On a curve, a chord gives an average rate and a tangent estimates a local one. Graph units determine whether the rate is speed, volume per time or another quantity; correlation alone does not prove causation.
- The three elapsed times sum to 120.0 s, giving mean 40.0 s and period 2.00 s for twenty oscillations. Range is 0.8 s; the chosen classroom half-range is 0.4 s for elapsed time and 0.020 s per period, one percent of the corresponding mean. This is not an automatic confidence interval or complete uncertainty budget. A separately stipulated fixed addition of 0.60 s is corrected by subtraction, giving mean 39.4 s and period 1.97 s without changing the repeat spread.
- Adding 44.0 s as another twenty-count run gives total 164.0 s over four, mean 41.0 s and period 2.05 s. Investigate timing, counting and conditions before excluding it. A genuine contemporaneous note of twenty-two oscillations would identify a non-equivalent count; it must be documented and cannot be pooled unchanged as a twenty-count run. Such a note must not be invented after seeing the desired answer. Repeating measurements does not automatically remove a shared bias.
- For the separate stipulated limits, minimum speed is 0.790 divided by 0.420, about 1.88095 m/s; maximum is 0.810 divided by 0.380, about 2.13158 m/s. The enclosing interval rounded outward to two decimal places is 1.88 to 2.14 m/s. The central value is 2.00 m/s. These are supplied allowed model bounds rather than probability-based confidence limits. Name omitted effects, raw-data handling and model scope; neither the pictures nor watching a video constitutes a performed practical.
Learning evidence
Keep dimensional conversions, the original force-extension gradients with units, both timing means, the declared half-range convention, the documented correction and positive-ratio limit calculation. Distinguish supplied model bounds from statistical uncertainty and actual source media from numerical observations.
OER and Wikipedia
Original unit conversions, force-extension models, repeated twenty-oscillation records, stipulated correction and hard speed limits. These are original educational cases, not measurements taken from the source photographs/videos.
DfE GCSE combined science, mathematical skills and working scientifically. BIPM SI Brochure and its 2026 concise summary, units and decimal prefixes. NIST TN1297 terminology and NIST measurement uncertainty, conceptual distinctions; the course’s explicitly chosen classroom half-range is not represented as a full NIST uncertainty calculation. Source reading checked 13 November 2026.
Additional original video: Science Buddies — The Scientific Method: Steps and Examples, metadata, preview and embedding verified. Original Commons bytes, credits and native render evidence retained.
Scale ruler, 1 foot and 300 mm.jpg, Kaviyadharshini Selvakumar, CC BY-SA 4.0. Unchanged original; caption and learning activities are original.
Stoppuhr.jpg, Ulfbastel, Public domain. Unchanged original; caption and learning activities are original.
2D Cartesian Coordinates.svg, H Padleckas, Public domain. Unchanged original; caption and learning activities are original.
Linked learning areas
aiMOOC Projects
Schulfach+


aiMOOCs


aiMOOC Projekte


MOOCwiki · Deutsch
Nach dem Lernen ist vor dem Lernen
Entdecke direkt den nächsten Lernkurs. Weitere Inhalte erscheinen, wenn Du weiter nach unten scrollst.
Zur MOOCwiki-HauptseiteMediathek
Mediathek wird aus dem Wiki geladen ...
Keine passenden Inhalte gefunden. Bitte ändere Suche oder Filter.
NEWSLernweltNOAH fragen