Why a Shopping Trolley Is a Geometry Problem on Wheels
A shopping trolley must roll straight, turn in narrow aisles, carry changing loads and remain stable over ordinary floors. Its casters swivel, its inner and outer wheels travel different distances, and the load's centre of mass moves as groceries are added. Mathematics turns these observations into path geometry, load sharing and stability diagrams.
Use drawings or a small unloaded model. Never ride, overload or deliberately tip a real trolley, and never conduct a test in a working shop or on a slope.
Caster Geometry Helps the Wheels Align
Swivel axis and trail
A typical swivel caster places the wheel contact behind the vertical swivel axis in the direction of travel. This offset, often called trail, creates a restoring moment that helps the wheel align with motion. The wheel swings through an arc around the swivel axis before rolling straight.
Suppose caster offset is 30 mm. A 90° swivel moves the axle centre along a quarter-circle of radius 30 mm, with arc length 30 × π/2 ≈ 47.1 mm. That small repositioning helps explain the initial sideways shuffle when a stationary trolley starts turning.
Caster designs differ. Some trolleys use four swivel casters; others constrain rear wheels. Draw the actual arrangement before applying a model.
Did You Know? A caster's wheel can rotate about its axle and the whole fork can rotate about the swivel axis. Those are two different rotations with different radii.
Inner and Outer Wheels Follow Different Paths
Turning-radius geometry
Model the trolley as a rectangle with track width 0.55 m. During a turn, suppose the midpoint follows a circle of radius 1.20 m. The inner path radius is about 1.20 − 0.275 = 0.925 m and outer radius is 1.475 m.
For a 90° turn, arc lengths are:
- inner: 0.925 × π/2 ≈ 1.45 m;
- midpoint: 1.20 × π/2 ≈ 1.88 m;
- outer: 1.475 × π/2 ≈ 2.32 m.
The outer side travels roughly 0.87 m farther than the inner side. Swivelling wheels and differential rolling accommodate this geometry without a powered differential.
| Path | Radius | Distance through 90° |
|---|---|---|
| Inner | 0.925 m | 1.45 m |
| Centre | 1.200 m | 1.88 m |
| Outer | 1.475 m | 2.32 m |
Corridor clearance
A trolley's corners sweep outside the centre path. A plan-view drawing can estimate clearance, but real steering includes caster delay and human correction. Add a margin rather than treating the geometric envelope as a promise.
Stability Depends on the Support Polygon
Join the wheel contact points on the floor. Their convex boundary is the support polygon. In a static model, the trolley remains upright while the vertical projection of the combined centre of mass stays inside this polygon.
If the projection reaches an edge, that edge becomes a tipping line. The angle can be modelled using horizontal half-width b and centre-of-mass height h: tan θ = b/h for a simple rigid body.
For b = 0.30 m and h = 0.60 m, θ = arctan(0.5) ≈ 26.6°. Raising the load centre to 0.90 m gives θ = arctan(0.30/0.90) ≈ 18.4°. A higher centre of mass reduces the static tilt angle in this simplified model.
This is not permission to test a real trolley. Moving loads, children, wheel stops, slopes and impacts can destabilise it well before an ideal calculation feels comfortable.
Load Position and Wheel Forces
Consider a simplified trolley supported at front and rear axles separated by 0.80 m. A 400 N load acts 0.30 m behind the front axle. Taking moments about the front axle gives rear reaction Rr × 0.80 = 400 × 0.30, so Rr = 150 N. Front reaction is 400 − 150 = 250 N.
Moving the load rearward changes these reactions. Equal division by four wheels is valid only for a symmetric, level and suitably distributed case. Frame flexibility and floor irregularity can make actual wheel loads unequal.
Push force and rolling resistance
On level ground, a simple model uses required steady push F ≈ CrrW, where Crr is a rolling-resistance coefficient and W is total weight. If W = 600 N and Crr = 0.02, F ≈ 12 N. Starting force may be larger because caster alignment, bearing friction and deformation must be overcome.
On a slope angle θ, gravity contributes W sin θ along the slope. Even a small angle can dominate rolling resistance. This is one reason real trolley use follows site rules rather than a classroom coefficient.
Standards and Safe Scope
The European EN 1929 family covers safety, construction, performance and testing for basket trolleys. A public catalogue summary of EN 1929-1:1998 describes general-purpose basket trolley requirements.
This article does not reproduce protected test procedures or invent load limits. It uses the existence of a broader standard to show why turning-radius and static-stability calculations are only parts of an evidence package.
Measurement and Uncertainty
Use a top-down photograph of a small model with a scale in the same plane. Measure wheelbase, track width and caster offset three times. If track width is 550 ± 5 mm, half-width is 275 ± 2.5 mm, affecting both turning-path and support-polygon calculations.
For path tracing, mark the model's centre and each corner on paper. Compare predicted arcs with traced paths. Residual differences reveal caster slip, steering corrections and measurement error.
Common Misconceptions
“All four wheels travel the same distance in a turn”
Inner and outer paths have different radii and arc lengths.
“Equal weight means equal load on every wheel”
Load position, frame geometry and floor contact determine reactions.
“A wide base can never tip”
Width helps, but centre-of-mass height, motion, slopes and impacts also matter.
“A static model proves a real trolley is safe”
It does not. Product safety requires applicable tests, instructions and maintenance.
How Students Can Learn the Mathematics
- Draw a plan view with wheelbase and track width.
- Trace caster and corner paths through a model turn.
- Calculate inner and outer arc lengths.
- Draw the support polygon and load projection.
- Solve a two-support moment problem.
- Plot safe supplied push-force data against load.
- State which effects the static model omits.
Parents can ask, “Where is the combined centre of mass now?” That question connects grocery placement to stability without encouraging risky testing.
Useful Next Reading
Compare public-flow mathematics in Why Mathematics? | Supermarket Checkouts, Queue Lengths and Little’s Law and explore the eduKate Mathematics Learning Hub.
Frequently Asked Questions
Why do trolley wheels swivel?
Swivel casters let wheel direction align with the trolley's changing motion.
Why does the trolley resist a sudden sideways start?
Casters may need to rotate into alignment before rolling smoothly.
What is a support polygon?
It is the boundary formed by floor contact points; static stability depends on the centre-of-mass projection remaining inside it.
Can students test tipping angles?
Not on real trolleys. Use diagrams, calculations or a lightweight scale model under supervision.
The Bigger Answer to “Why Mathematics?”
A shopping trolley shows geometry in motion. Arc lengths separate inner and outer paths, moments distribute loads, polygons describe stability and uncertainty identifies the limits. Mathematics helps explain a familiar machine while preserving the gap between a model and a safety decision.
A Practical Mathematics Studio
Use drawings or a small unloaded model in a clear indoor area. Never climb into, ride, overload or deliberately tip a shopping trolley, and do not obstruct shoppers, slopes, travelators or vehicle routes. Keep raw observations, units, assumptions and limitations beside each result.
Investigation 1: Draw the trolley footprint
Declare footprint, wheel arrangement, caster offset, load position, surface and direction of travel. Draw the relevant path or support polygon, calculate the named radius, load or stability quantity, and separate static models from moving behaviour. Use a diagram first route for this investigation so the evidence is not reduced to a single answer.
Evidence check: Check caster orientation, inside versus outside paths, support polygon and load height. A static tipping line does not include acceleration, bumps, slopes or a moving child. For draw the trolley footprint, repeat or independently recompute at least one result, then compare it with a second representation before drawing a conclusion.
Boundary: The exercise cannot certify a trolley, child seat or travelator use. Obey the retailer’s instructions, marked loads and applicable safety requirements. End with one sentence stating what this particular investigation supports and one naming the evidence still missing.
Investigation 2: Mark wheelbase and track width
Declare footprint, wheel arrangement, caster offset, load position, surface and direction of travel. Draw the relevant path or support polygon, calculate the named radius, load or stability quantity, and separate static models from moving behaviour. Use a table first route for this investigation so the evidence is not reduced to a single answer.
Evidence check: Check caster orientation, inside versus outside paths, support polygon and load height. A static tipping line does not include acceleration, bumps, slopes or a moving child. For mark wheelbase and track width, repeat or independently recompute at least one result, then compare it with a second representation before drawing a conclusion.
Boundary: The exercise cannot certify a trolley, child seat or travelator use. Obey the retailer’s instructions, marked loads and applicable safety requirements. End with one sentence stating what this particular investigation supports and one naming the evidence still missing.
Investigation 3: Identify swivel and rigid casters
Declare footprint, wheel arrangement, caster offset, load position, surface and direction of travel. Draw the relevant path or support polygon, calculate the named radius, load or stability quantity, and separate static models from moving behaviour. Use a equation first route for this investigation so the evidence is not reduced to a single answer.
Evidence check: Check caster orientation, inside versus outside paths, support polygon and load height. A static tipping line does not include acceleration, bumps, slopes or a moving child. For identify swivel and rigid casters, repeat or independently recompute at least one result, then compare it with a second representation before drawing a conclusion.
Boundary: The exercise cannot certify a trolley, child seat or travelator use. Obey the retailer’s instructions, marked loads and applicable safety requirements. End with one sentence stating what this particular investigation supports and one naming the evidence still missing.
Investigation 4: Measure caster offset on a model
Declare footprint, wheel arrangement, caster offset, load position, surface and direction of travel. Draw the relevant path or support polygon, calculate the named radius, load or stability quantity, and separate static models from moving behaviour. Use a graph first route for this investigation so the evidence is not reduced to a single answer.
Evidence check: Check caster orientation, inside versus outside paths, support polygon and load height. A static tipping line does not include acceleration, bumps, slopes or a moving child. For measure caster offset on a model, repeat or independently recompute at least one result, then compare it with a second representation before drawing a conclusion.
Boundary: The exercise cannot certify a trolley, child seat or travelator use. Obey the retailer’s instructions, marked loads and applicable safety requirements. End with one sentence stating what this particular investigation supports and one naming the evidence still missing.
Investigation 5: Trace a wheel path through a turn
Declare footprint, wheel arrangement, caster offset, load position, surface and direction of travel. Draw the relevant path or support polygon, calculate the named radius, load or stability quantity, and separate static models from moving behaviour. Use a comparison first route for this investigation so the evidence is not reduced to a single answer.
Evidence check: Check caster orientation, inside versus outside paths, support polygon and load height. A static tipping line does not include acceleration, bumps, slopes or a moving child. For trace a wheel path through a turn, repeat or independently recompute at least one result, then compare it with a second representation before drawing a conclusion.
Boundary: The exercise cannot certify a trolley, child seat or travelator use. Obey the retailer’s instructions, marked loads and applicable safety requirements. End with one sentence stating what this particular investigation supports and one naming the evidence still missing.
Investigation 6: Estimate inside turning radius
Declare footprint, wheel arrangement, caster offset, load position, surface and direction of travel. Draw the relevant path or support polygon, calculate the named radius, load or stability quantity, and separate static models from moving behaviour. Use a sensitivity first route for this investigation so the evidence is not reduced to a single answer.
Evidence check: Check caster orientation, inside versus outside paths, support polygon and load height. A static tipping line does not include acceleration, bumps, slopes or a moving child. For estimate inside turning radius, repeat or independently recompute at least one result, then compare it with a second representation before drawing a conclusion.
Boundary: The exercise cannot certify a trolley, child seat or travelator use. Obey the retailer’s instructions, marked loads and applicable safety requirements. End with one sentence stating what this particular investigation supports and one naming the evidence still missing.
Investigation 7: Compare inner and outer wheel distance
Declare footprint, wheel arrangement, caster offset, load position, surface and direction of travel. Draw the relevant path or support polygon, calculate the named radius, load or stability quantity, and separate static models from moving behaviour. Use a diagram first route for this investigation so the evidence is not reduced to a single answer.
Evidence check: Check caster orientation, inside versus outside paths, support polygon and load height. A static tipping line does not include acceleration, bumps, slopes or a moving child. For compare inner and outer wheel distance, repeat or independently recompute at least one result, then compare it with a second representation before drawing a conclusion.
Boundary: The exercise cannot certify a trolley, child seat or travelator use. Obey the retailer’s instructions, marked loads and applicable safety requirements. End with one sentence stating what this particular investigation supports and one naming the evidence still missing.
Investigation 8: Calculate distance ratio
Declare footprint, wheel arrangement, caster offset, load position, surface and direction of travel. Draw the relevant path or support polygon, calculate the named radius, load or stability quantity, and separate static models from moving behaviour. Use a table first route for this investigation so the evidence is not reduced to a single answer.
Evidence check: Check caster orientation, inside versus outside paths, support polygon and load height. A static tipping line does not include acceleration, bumps, slopes or a moving child. For calculate distance ratio, repeat or independently recompute at least one result, then compare it with a second representation before drawing a conclusion.
Boundary: The exercise cannot certify a trolley, child seat or travelator use. Obey the retailer’s instructions, marked loads and applicable safety requirements. End with one sentence stating what this particular investigation supports and one naming the evidence still missing.
Investigation 9: Model a corridor clearance
Declare footprint, wheel arrangement, caster offset, load position, surface and direction of travel. Draw the relevant path or support polygon, calculate the named radius, load or stability quantity, and separate static models from moving behaviour. Use a equation first route for this investigation so the evidence is not reduced to a single answer.
Evidence check: Check caster orientation, inside versus outside paths, support polygon and load height. A static tipping line does not include acceleration, bumps, slopes or a moving child. For model a corridor clearance, repeat or independently recompute at least one result, then compare it with a second representation before drawing a conclusion.
Boundary: The exercise cannot certify a trolley, child seat or travelator use. Obey the retailer’s instructions, marked loads and applicable safety requirements. End with one sentence stating what this particular investigation supports and one naming the evidence still missing.
Investigation 10: Place a model centre of mass
Declare footprint, wheel arrangement, caster offset, load position, surface and direction of travel. Draw the relevant path or support polygon, calculate the named radius, load or stability quantity, and separate static models from moving behaviour. Use a graph first route for this investigation so the evidence is not reduced to a single answer.
Evidence check: Check caster orientation, inside versus outside paths, support polygon and load height. A static tipping line does not include acceleration, bumps, slopes or a moving child. For place a model centre of mass, repeat or independently recompute at least one result, then compare it with a second representation before drawing a conclusion.
Boundary: The exercise cannot certify a trolley, child seat or travelator use. Obey the retailer’s instructions, marked loads and applicable safety requirements. End with one sentence stating what this particular investigation supports and one naming the evidence still missing.
Investigation 11: Draw the support polygon
Declare footprint, wheel arrangement, caster offset, load position, surface and direction of travel. Draw the relevant path or support polygon, calculate the named radius, load or stability quantity, and separate static models from moving behaviour. Use a comparison first route for this investigation so the evidence is not reduced to a single answer.
Evidence check: Check caster orientation, inside versus outside paths, support polygon and load height. A static tipping line does not include acceleration, bumps, slopes or a moving child. For draw the support polygon, repeat or independently recompute at least one result, then compare it with a second representation before drawing a conclusion.
Boundary: The exercise cannot certify a trolley, child seat or travelator use. Obey the retailer’s instructions, marked loads and applicable safety requirements. End with one sentence stating what this particular investigation supports and one naming the evidence still missing.
Investigation 12: Test a gentle static tilt model
Declare footprint, wheel arrangement, caster offset, load position, surface and direction of travel. Draw the relevant path or support polygon, calculate the named radius, load or stability quantity, and separate static models from moving behaviour. Use a sensitivity first route for this investigation so the evidence is not reduced to a single answer.
Evidence check: Check caster orientation, inside versus outside paths, support polygon and load height. A static tipping line does not include acceleration, bumps, slopes or a moving child. For test a gentle static tilt model, repeat or independently recompute at least one result, then compare it with a second representation before drawing a conclusion.
Boundary: The exercise cannot certify a trolley, child seat or travelator use. Obey the retailer’s instructions, marked loads and applicable safety requirements. End with one sentence stating what this particular investigation supports and one naming the evidence still missing.
Investigation 13: Estimate a tipping threshold
Declare footprint, wheel arrangement, caster offset, load position, surface and direction of travel. Draw the relevant path or support polygon, calculate the named radius, load or stability quantity, and separate static models from moving behaviour. Use a diagram first route for this investigation so the evidence is not reduced to a single answer.
Evidence check: Check caster orientation, inside versus outside paths, support polygon and load height. A static tipping line does not include acceleration, bumps, slopes or a moving child. For estimate a tipping threshold, repeat or independently recompute at least one result, then compare it with a second representation before drawing a conclusion.
Boundary: The exercise cannot certify a trolley, child seat or travelator use. Obey the retailer’s instructions, marked loads and applicable safety requirements. End with one sentence stating what this particular investigation supports and one naming the evidence still missing.
Investigation 14: Move load forward and backward
Declare footprint, wheel arrangement, caster offset, load position, surface and direction of travel. Draw the relevant path or support polygon, calculate the named radius, load or stability quantity, and separate static models from moving behaviour. Use a table first route for this investigation so the evidence is not reduced to a single answer.
Evidence check: Check caster orientation, inside versus outside paths, support polygon and load height. A static tipping line does not include acceleration, bumps, slopes or a moving child. For move load forward and backward, repeat or independently recompute at least one result, then compare it with a second representation before drawing a conclusion.
Boundary: The exercise cannot certify a trolley, child seat or travelator use. Obey the retailer’s instructions, marked loads and applicable safety requirements. End with one sentence stating what this particular investigation supports and one naming the evidence still missing.
Investigation 15: Compare balanced and unbalanced loads
Declare footprint, wheel arrangement, caster offset, load position, surface and direction of travel. Draw the relevant path or support polygon, calculate the named radius, load or stability quantity, and separate static models from moving behaviour. Use a equation first route for this investigation so the evidence is not reduced to a single answer.
Evidence check: Check caster orientation, inside versus outside paths, support polygon and load height. A static tipping line does not include acceleration, bumps, slopes or a moving child. For compare balanced and unbalanced loads, repeat or independently recompute at least one result, then compare it with a second representation before drawing a conclusion.
Boundary: The exercise cannot certify a trolley, child seat or travelator use. Obey the retailer’s instructions, marked loads and applicable safety requirements. End with one sentence stating what this particular investigation supports and one naming the evidence still missing.
Investigation 16: Calculate total wheel load
Declare footprint, wheel arrangement, caster offset, load position, surface and direction of travel. Draw the relevant path or support polygon, calculate the named radius, load or stability quantity, and separate static models from moving behaviour. Use a graph first route for this investigation so the evidence is not reduced to a single answer.
Evidence check: Check caster orientation, inside versus outside paths, support polygon and load height. A static tipping line does not include acceleration, bumps, slopes or a moving child. For calculate total wheel load, repeat or independently recompute at least one result, then compare it with a second representation before drawing a conclusion.
Boundary: The exercise cannot certify a trolley, child seat or travelator use. Obey the retailer’s instructions, marked loads and applicable safety requirements. End with one sentence stating what this particular investigation supports and one naming the evidence still missing.
Investigation 17: Estimate equal-load share
Declare footprint, wheel arrangement, caster offset, load position, surface and direction of travel. Draw the relevant path or support polygon, calculate the named radius, load or stability quantity, and separate static models from moving behaviour. Use a comparison first route for this investigation so the evidence is not reduced to a single answer.
Evidence check: Check caster orientation, inside versus outside paths, support polygon and load height. A static tipping line does not include acceleration, bumps, slopes or a moving child. For estimate equal-load share, repeat or independently recompute at least one result, then compare it with a second representation before drawing a conclusion.
Boundary: The exercise cannot certify a trolley, child seat or travelator use. Obey the retailer’s instructions, marked loads and applicable safety requirements. End with one sentence stating what this particular investigation supports and one naming the evidence still missing.
Investigation 18: Discuss dynamic load transfer
Declare footprint, wheel arrangement, caster offset, load position, surface and direction of travel. Draw the relevant path or support polygon, calculate the named radius, load or stability quantity, and separate static models from moving behaviour. Use a sensitivity first route for this investigation so the evidence is not reduced to a single answer.
Evidence check: Check caster orientation, inside versus outside paths, support polygon and load height. A static tipping line does not include acceleration, bumps, slopes or a moving child. For discuss dynamic load transfer, repeat or independently recompute at least one result, then compare it with a second representation before drawing a conclusion.
Boundary: The exercise cannot certify a trolley, child seat or travelator use. Obey the retailer’s instructions, marked loads and applicable safety requirements. End with one sentence stating what this particular investigation supports and one naming the evidence still missing.
Investigation 19: Record push force on level ground
Declare footprint, wheel arrangement, caster offset, load position, surface and direction of travel. Draw the relevant path or support polygon, calculate the named radius, load or stability quantity, and separate static models from moving behaviour. Use a diagram first route for this investigation so the evidence is not reduced to a single answer.
Evidence check: Check caster orientation, inside versus outside paths, support polygon and load height. A static tipping line does not include acceleration, bumps, slopes or a moving child. For record push force on level ground, repeat or independently recompute at least one result, then compare it with a second representation before drawing a conclusion.
Boundary: The exercise cannot certify a trolley, child seat or travelator use. Obey the retailer’s instructions, marked loads and applicable safety requirements. End with one sentence stating what this particular investigation supports and one naming the evidence still missing.
Investigation 20: Plot force against model load
Declare footprint, wheel arrangement, caster offset, load position, surface and direction of travel. Draw the relevant path or support polygon, calculate the named radius, load or stability quantity, and separate static models from moving behaviour. Use a table first route for this investigation so the evidence is not reduced to a single answer.
Evidence check: Check caster orientation, inside versus outside paths, support polygon and load height. A static tipping line does not include acceleration, bumps, slopes or a moving child. For plot force against model load, repeat or independently recompute at least one result, then compare it with a second representation before drawing a conclusion.
Boundary: The exercise cannot certify a trolley, child seat or travelator use. Obey the retailer’s instructions, marked loads and applicable safety requirements. End with one sentence stating what this particular investigation supports and one naming the evidence still missing.
Investigation 21: Inspect caster alignment delay
Declare footprint, wheel arrangement, caster offset, load position, surface and direction of travel. Draw the relevant path or support polygon, calculate the named radius, load or stability quantity, and separate static models from moving behaviour. Use a equation first route for this investigation so the evidence is not reduced to a single answer.
Evidence check: Check caster orientation, inside versus outside paths, support polygon and load height. A static tipping line does not include acceleration, bumps, slopes or a moving child. For inspect caster alignment delay, repeat or independently recompute at least one result, then compare it with a second representation before drawing a conclusion.
Boundary: The exercise cannot certify a trolley, child seat or travelator use. Obey the retailer’s instructions, marked loads and applicable safety requirements. End with one sentence stating what this particular investigation supports and one naming the evidence still missing.
Investigation 22: Check slope sensitivity
Declare footprint, wheel arrangement, caster offset, load position, surface and direction of travel. Draw the relevant path or support polygon, calculate the named radius, load or stability quantity, and separate static models from moving behaviour. Use a graph first route for this investigation so the evidence is not reduced to a single answer.
Evidence check: Check caster orientation, inside versus outside paths, support polygon and load height. A static tipping line does not include acceleration, bumps, slopes or a moving child. For check slope sensitivity, repeat or independently recompute at least one result, then compare it with a second representation before drawing a conclusion.
Boundary: The exercise cannot certify a trolley, child seat or travelator use. Obey the retailer’s instructions, marked loads and applicable safety requirements. End with one sentence stating what this particular investigation supports and one naming the evidence still missing.
Investigation 23: Read the trolley-standard scope
Declare footprint, wheel arrangement, caster offset, load position, surface and direction of travel. Draw the relevant path or support polygon, calculate the named radius, load or stability quantity, and separate static models from moving behaviour. Use a comparison first route for this investigation so the evidence is not reduced to a single answer.
Evidence check: Check caster orientation, inside versus outside paths, support polygon and load height. A static tipping line does not include acceleration, bumps, slopes or a moving child. For read the trolley-standard scope, repeat or independently recompute at least one result, then compare it with a second representation before drawing a conclusion.
Boundary: The exercise cannot certify a trolley, child seat or travelator use. Obey the retailer’s instructions, marked loads and applicable safety requirements. End with one sentence stating what this particular investigation supports and one naming the evidence still missing.
Investigation 24: State a bounded conclusion
Declare footprint, wheel arrangement, caster offset, load position, surface and direction of travel. Draw the relevant path or support polygon, calculate the named radius, load or stability quantity, and separate static models from moving behaviour. Use a sensitivity first route for this investigation so the evidence is not reduced to a single answer.
Evidence check: Check caster orientation, inside versus outside paths, support polygon and load height. A static tipping line does not include acceleration, bumps, slopes or a moving child. For state a bounded conclusion, repeat or independently recompute at least one result, then compare it with a second representation before drawing a conclusion.
Boundary: The exercise cannot certify a trolley, child seat or travelator use. Obey the retailer’s instructions, marked loads and applicable safety requirements. End with one sentence stating what this particular investigation supports and one naming the evidence still missing.
Studio Synthesis
Choose four investigations that use different representations: a labelled system diagram, an equation, a raw-data table and a graph. Arrange the report from question to method, calculation, residual or sensitivity check, limitations and conclusion. Another reader should be able to reproduce one result without guessing a unit, category or selection rule.
Change one input or classification rule at a time and recalculate. Identify which conclusion remains stable and which changes direction. Sensitivity analysis reveals the assumptions controlling a result and is usually more valuable than an extra decimal place.
Exchange only the raw data and definitions with a partner. If the calculation reproduces and the same bounded conclusion follows, confidence increases. If it does not, repair the method or labels before treating the output as evidence.