eduKate SG

Three learners review open books together at a classroom table, with stacks of textbooks, stationery and a whiteboard in the bright room.

eduKate Learning Manual: Graph Linearisation and Model Testing | Turning Curves Into Evidence Without Forcing the Theory

Wait, What? Making a graph straight does not prove the theory is right.

Almost any small dataset can be made to look persuasive if axes, transformations and fit choices are selected after seeing the answer. Scientific graphing works in the opposite direction: start from the model, derive the predicted relationship, choose a transformation that tests it, and then allow the data to disagree.

The RFE: a graph is a model test, not decoration

A graph should answer a scientific question. Is y proportional to x? Does y vary as x²? Is there an inverse relationship? Does an exponential law fit? The axes are chosen because the model predicts a particular structure.

Direct proportionality

If y = kx, plotting y against x should give a straight line through the origin with gradient k—within uncertainty and model limits.

A straight line with a large non-zero intercept is not direct proportionality. It may indicate an offset, background, friction, dead volume or a missing term in the model.

Power relationships

If y = kx², plotting y against x gives a curve, but plotting y against x² should give a straight line.

If y = k/x, plotting y against 1/x should be linear. This is the logic behind Boyle-type p against 1/V plots and many other school transformations.

Linearisation must come from algebra

Do not try x, x², √x and 1/x until one looks straight. Derive the transformation from the physical equation first. Otherwise the graph becomes a search for appearance rather than a test of prediction.

Quantitative window: spring oscillation

For a mass–spring oscillator:

T = 2π√(m/k)

Squaring gives:

T² = (4π²/k)m

Therefore plot T² vertically against m horizontally. The gradient is 4π²/k, so k can be inferred. A non-zero intercept can reveal effective spring mass or timing/model offsets.

Gradient has physical meaning

A gradient is not just “rise over run.” Its units and physical interpretation come from the equation. If a graph of V against I has gradient R, the gradient units are V A⁻¹ = Ω. If T² against m has gradient 4π²/k, its units must be s² kg⁻¹.

Checking gradient units is one of the fastest ways to catch a wrong axis choice.

Intercepts are evidence

Students often call every intercept “experimental error.” That throws away information. An intercept may represent zero error, background signal, friction threshold, dead volume, effective mass or a real constant term in the model.

Ask what physical quantity the intercept would represent if the fitted equation is correct.

Do not force the line through the origin

If theory predicts zero intercept but data do not, an unconstrained fit lets the experiment reveal the discrepancy. Forcing the line through zero can change the gradient and hide systematic error.

Logarithms can test general power laws

If y = Axⁿ, then:

ln y = ln A + n ln x

A plot of ln y against ln x can therefore have gradient n. This is useful at higher levels, but logarithms require positive quantities and transform uncertainty too. They should not be used merely because software offers the button.

Straightness is not enough

A dataset with only three points can look perfectly linear while providing weak discrimination between models. Use enough range and enough points to reveal curvature. Range finding and pilot trials help choose values that actually test the relationship.

Error bars and model compatibility

Where appropriate, uncertainty bars show whether apparent deviations are large compared with measurement uncertainty. A model line passing close to all central points is less impressive if the uncertainty bars are enormous; conversely, a small visible deviation can matter when measurements are very precise.

Observation versus inference

Observation: “T² plotted against m is approximately linear, with a positive intercept.”

Inference: “The data support the predicted square-root period dependence over this range, while the intercept suggests an additional effective-mass or systematic contribution.”

Overclaim: “The straight line proves simple harmonic motion exactly.”

Failure modes

Unfamiliar transfer: inverse-square behaviour

If a signal follows S = k/r², plotting S against r is curved. Plot S against 1/r² to test the predicted proportionality. But if a background B exists, the better model is S = k/r² + B, and the intercept becomes physically meaningful.

Secondary → JC → deeper Science

Secondary: plot independent and dependent variables correctly, draw best-fit trends and interpret gradient qualitatively.

JC: derive linearising transformations, interpret gradients/intercepts quantitatively and use uncertainty to test models.

Deeper Science: extend to nonlinear regression, likelihood, model selection, parameter covariance and cross-validation.

Checkpoint

The model predicts y = kx². Which graph should be linear, and what should its gradient represent?

Answer key and WHY reasoning

Plot y against x². The gradient should be k. This follows directly from rewriting the equation as y = k(x²), so the transformed horizontal variable is x².

How to study this

Use the sequence model equation → rearrangement → predicted axes → gradient meaning → intercept meaning → residual pattern. Never begin with graph-paper aesthetics.

Evidence boundaries

Linearisation can make a predicted relationship easier to test, but transformations can change uncertainty structure and visual weighting. A straight transformed plot supports a model over the measured range; it does not establish universal truth or unique mechanism.

Authoritative next steps

Teaching Guide

Give students one curved dataset and three possible models. Require them to derive each predicted transformation before plotting anything. The learning target is not making a straight line; it is allowing the model to risk being wrong.

Explore the connected learning guides

Choose the question that brought you here. Open one useful guide, try a small task, and stop when you have what you need.

Take one question further

The same learning habit can travel across subjects, while each subject keeps its own methods. These routes help you notice a difficulty, understand one part of it, and return to something you can do.

A word is familiar, but using it is difficult.

Move from recognising a word to retrieving it in a new context. Understand vocabulary plateaus.

Try it without the guide: Choose one word you already know. Close the guide and use it in a new sentence. Explain why it fits; try another context tomorrow.

A piece of writing has ideas, but the reader loses the thread.

Make the order of events and the links between sentences clear. Explore composition writing.

Try it without the guide: Choose one short paragraph. Read the relevant explanation, close it, and revise the paragraph. Ask someone to tell you what happened and why.

The Mathematics seems familiar, but marks still disappear.

Find the first point where the working stops being reliable. Find Secondary 4 A-Math mark leakage.

Try it without the guide: For a Secondary 4 A-Math question you have attempted, locate the first uncertain line. Repair that step, then try a comparable question without the worked answer.

A Science fact is remembered, but the explanation is incomplete.

Connect the evidence to a scientific idea and the resulting change. Follow the Primary Science learning route.

Try it without the guide: Choose a familiar Primary Science example. Explain the evidence, the idea and the result without notes. Then change one condition and explain your prediction.

Two accounts of the world seem to disagree.

Check the question, source, date and evidence before combining claims. Explore the World Knowledge research library.

Try it without the guide: Take one claim. Find the source best placed to support it, note its date, and state what remains uncertain. Return to your original question.

There is plenty of help, but independence is hard to see.

Check what the learner can understand and do after support is removed. Understand how education works.

Try it without the guide: Choose one small task the child has practised. Agree on a calm, brief attempt without prompts. Use what happens to choose one next step, then stop.

For the structure behind these connections, read the eduKateSingapore runtime manifest and the eduKate ecosystem boot contract. The reader map describes public navigation; those manifests preserve the wider ownership and return rules.

eduKate Asia

Subscribe to
our newsletter

Discover more from eduKate SG

Subscribe now to keep reading and get access to the full archive.

Continue reading