Skip to main content
SingaporePhysicsSyllabus dot point

How do we linearise a physical relationship so that a straight-line graph yields the quantities we want from its gradient and intercept?

Rearrange a physical relationship into straight-line form y = mx + c, plot the appropriate variables, and extract physical quantities from the gradient and intercept

A focused answer to the H2 Physics Measurement learning outcome on graphical analysis. Linearising relationships into y = mx + c form, choosing axes, and reading physical quantities from gradient and intercept in Paper 3 and Paper 4.

Generated by Claude Opus 4.88 min answer

Reviewed by: AI editorial process; not yet individually human-reviewed

Have a quick question? Jump to the Q&A page

Jump to a section
  1. What this dot point is asking
  2. The answer
  3. Examples in context
  4. Try this

What this dot point is asking

SEAB wants you to take a physical relationship that may be non-linear, rearrange it into the straight-line form y=mx+cy = mx + c, decide what to plot on each axis, and then extract the physical quantity you want from the gradient and the intercept. This skill is central to the Paper 3 data-based question and to the analysis section of the Paper 4 practical.

The answer

Why a straight line is the goal

A straight line is the easiest graph to interpret: its gradient and intercept are read directly and a best-fit line averages out random scatter. So the strategy for any relationship is to manipulate it into the form:

y=mx+cy = mx + c

where mm is the gradient and cc is the yy-intercept, then plot the combinations that play the roles of yy and xx.

Linearising common relationships

The key is to identify which grouping of variables behaves like yy, which like xx, and what the gradient and intercept then represent.

  • v=u+atv = u + at is already linear: plot vv against tt; gradient =a= a, intercept =u= u.
  • T2=4π2gLT^2 = \dfrac{4\pi^2}{g} L for a pendulum: plot T2T^2 against LL; gradient =4π2g= \dfrac{4\pi^2}{g}.
  • 1v+1u=1f\dfrac{1}{v} + \dfrac{1}{u} = \dfrac{1}{f} for a lens: plot 1v\dfrac{1}{v} against 1u\dfrac{1}{u}; intercept =1f= \dfrac{1}{f}.

Power laws and the log-log plot

If y=axny = a x^n, taking logarithms gives:

lgy=nlgx+lga\lg y = n \lg x + \lg a

Plot lgy\lg y against lgx\lg x. The gradient is the power nn and the intercept is lga\lg a, so a=10intercepta = 10^{\text{intercept}}. A straight log-log line confirms the power-law form.

Exponentials and the log-linear plot

If y=y0ekxy = y_0 e^{-kx} (radioactive decay, capacitor discharge), take natural logs:

lny=lny0kx\ln y = \ln y_0 - kx

Plot lny\ln y against xx. The gradient is k-k and the intercept is lny0\ln y_0. This is the standard way to find a decay or time constant from data.

Reading the gradient correctly

The gradient must be calculated from two well-separated points on the best-fit line, not from a single data point:

gradient=y2y1x2x1\text{gradient} = \frac{y_2 - y_1}{x_2 - x_1}

Always carry units through the gradient, because the physical quantity often comes from the gradient's units as much as its value.

Examples in context

Example 1. Finding gg from a pendulum. Rearranging T=2πL/gT = 2\pi\sqrt{L/g} to T2=4π2gLT^2 = \dfrac{4\pi^2}{g}L and plotting T2T^2 against LL gives a straight line through the origin. Measuring its gradient GG yields g=4π2Gg = \dfrac{4\pi^2}{G}. The straight-line method averages many trials and exposes any systematic offset as a non-zero intercept.

Example 2. Confirming an inverse-square field. Suspecting I1d2I \propto \dfrac{1}{d^2} for a point source, a student plots II against 1d2\dfrac{1}{d^2}. A straight line through the origin confirms the inverse-square law, while a log-log plot of lgI\lg I against lgd\lg d with gradient 2-2 provides independent confirmation of the power.

Try this

Q1. The relationship Ek=12mv2E_k = \tfrac{1}{2}mv^2 is to be tested by varying vv. State what you would plot to obtain a straight line through the origin, and what the gradient represents. [2 marks]

  • Cue. Plot EkE_k against v2v^2; gradient =12m= \tfrac{1}{2}m.

Q2. A set of resistance values follows R=R0(1+αθ)R = R_0(1 + \alpha\theta) with temperature θ\theta. State the axes for a straight-line plot and identify the gradient and intercept. [2 marks]

  • Cue. Plot RR against θ\theta; intercept =R0= R_0, gradient =R0α= R_0\alpha, so α=gradientintercept\alpha = \dfrac{\text{gradient}}{\text{intercept}}.

Q3. Data are suspected to follow y=ax3y = ax^3. Describe a logarithmic plot to confirm this and find aa. [3 marks]

  • Cue. Plot lgy\lg y against lgx\lg x; a straight line of gradient 33 confirms the cube law, and a=10intercepta = 10^{\text{intercept}}.

Exam-style practice questions

Practice questions written in the style of SEAB exam questions on this dot point, with worked answer explainers. The year tag is the paper they imitate, not the source.

Original5 marksA spring obeys T=2πmkT = 2\pi\sqrt{\dfrac{m}{k}}, where TT is the period of oscillation of a mass mm and kk is the spring constant. (a) Show how to plot a straight-line graph to determine kk. (b) State what you would plot on each axis and how kk is found from the gradient.
Show worked answer →

(a) Square both sides to remove the root: T2=4π2kmT^2 = \dfrac{4\pi^2}{k} m.

This is in the form y=mx+cy = mx + c with y=T2y = T^2, x=mx = m, gradient =4π2k= \dfrac{4\pi^2}{k} and intercept =0= 0.

(b) Plot T2T^2 (vertical axis) against mm (horizontal axis). The graph is a straight line through the origin with gradient 4π2k\dfrac{4\pi^2}{k}.

Rearranging: k=4π2gradientk = \dfrac{4\pi^2}{\text{gradient}}.

Markers reward the squaring step, the clear identification of which variable is yy and which is xx, the correct gradient expression, and the final rearrangement for kk.

Original4 marksThe current-voltage relationship for a component is suspected to be I=aVnI = aV^n. Explain how a logarithmic plot can test this relationship and find both aa and nn.
Show worked answer →

Take logarithms of both sides: lgI=lga+nlgV\lg I = \lg a + n \lg V.

This is in the form y=mx+cy = mx + c with y=lgIy = \lg I, x=lgVx = \lg V, gradient =n= n and intercept =lga= \lg a.

Plot lgI\lg I against lgV\lg V. If the data fall on a straight line, the power-law relationship holds. The gradient gives nn directly, and the constant is found from a=10intercepta = 10^{\text{intercept}}.

Markers reward taking logs of both sides, identifying the gradient as nn and the intercept as lga\lg a, and recovering aa by taking the antilog of the intercept.

Related dot points