Skip to content
spmaddmath.com.my
Tuition

Study

SyllabusFormulasMethodsExam & PapersTools
LocationsPricingBlogOur TeachersContact
EN

Calculus clinic · Differentiation

Confusing dy/dx with the second derivative

dydx\frac{dy}{dx} is the gradient of the curve at a point; d2ydx2\frac{d^{2}y}{dx^{2}} is the rate of change of that gradient, found by differentiating dydx\frac{dy}{dx} a second time. Mixing the two up makes the second-derivative test meaningless, because dydx=0\frac{dy}{dx}=0 at every stationary point by definition, only d2ydx2\frac{d^{2}y}{dx^{2}} can tell a maximum from a minimum.

The error

dydx\frac{dy}{dx} and d2ydx2\frac{d^{2}y}{dx^{2}} look almost identical on the page, the same fraction shape, the same dxdx underneath, but they answer different questions. dydx\frac{dy}{dx} is the first derivative: the gradient of the curve at a point, how fast yy is changing as xx changes.

d2ydx2\frac{d^{2}y}{dx^{2}} is the second derivative: the rate of change of that gradient itself, found by differentiating dydx\frac{dy}{dx} a second time. It tells you whether the gradient is increasing or decreasing, which is another way of describing whether the curve is concave up or concave down.

The confusion shows up in three places. A question that asks for the 'rate of change of the gradient' or the 'concavity' of a curve is asking for d2ydx2\frac{d^{2}y}{dx^{2}}, and stopping after one differentiation leaves the answer incomplete.

Classifying a stationary point with the second-derivative test but substituting the value of dydx\frac{dy}{dx} instead of d2ydx2\frac{d^{2}y}{dx^{2}} is worse than incomplete, it is meaningless, because dydx=0\frac{dy}{dx}=0 is exactly what makes the point stationary, so that value can never distinguish a maximum from a minimum. And in kinematics, reporting the velocity dsdt\frac{ds}{dt} when a question asks for the acceleration d2sdt2\frac{d^{2}s}{dt^{2}} is the same slip wearing different letters.

Wrong line, then the fix

A curve has equation y=2x39x2+12x+1y=2x^{3}-9x^{2}+12x+1. Find the rate of change of the gradient of the curve at x=0x=0.

First find the gradient function by differentiating once: dydx=6x218x+12\frac{dy}{dx}=6x^{2}-18x+12. At x=0x=0, this gives dydx=12\frac{dy}{dx}=12.

Wrong, this is the gradient itself, not its rate of changeMust memorise
dydxx=0=12  rate of change of the gradient=12\left.\frac{dy}{dx}\right|_{x=0}=12\ \Rightarrow\ \text{rate of change of the gradient}=12

The gradient at x=0x=0 really is 1212, but the question does not ask for the gradient. It asks for the rate of change of the gradient, which means differentiating the gradient function dydx\frac{dy}{dx} a second time.

Correct, differentiate dy/dx again to get d²y/dx²Must memorise
d2ydx2=12x18d2ydx2x=0=18\frac{d^{2}y}{dx^{2}}=12x-18\quad\Rightarrow\quad \left.\frac{d^{2}y}{dx^{2}}\right|_{x=0}=-18

So the rate of change of the gradient at x=0x=0 is 18-18, not 1212. The negative sign is part of the answer: it says the gradient is decreasing at that point, so the curve is bending over, concave down, there, which the first-derivative value of 1212 cannot tell you by itself.

The reliable fix

  1. 1

    Read the words before you differentiate

    'Gradient' or 'slope of the tangent' means dydx\frac{dy}{dx}. 'Rate of change of the gradient', 'concavity' or 'acceleration' means d2ydx2\frac{d^{2}y}{dx^{2}} (or d2sdt2\frac{d^{2}s}{dt^{2}} in kinematics).

    Decide which one the question wants before you touch a derivative.

  2. 2

    Differentiate twice for the second derivative

    d2ydx2\frac{d^{2}y}{dx^{2}} is not a separate rule, it is dydx\frac{dy}{dx}, differentiated again. Work out the first derivative fully, then differentiate that expression a second time.

  3. 3

    Never test a stationary point using dy/dx

    At a stationary point dydx=0\frac{dy}{dx}=0 by definition, so substituting it into the second-derivative test always gives 00, whatever the point's real nature. Substitute the xx-value into d2ydx2\frac{d^{2}y}{dx^{2}} instead.

  4. 4

    Keep displacement, velocity and acceleration in order

    ss\to differentiate once v=dsdt\to v=\frac{ds}{dt} (velocity) \to differentiate again a=d2sdt2=dvdt\to a=\frac{d^{2}s}{dt^{2}}=\frac{dv}{dt} (acceleration). Reporting vv where aa is asked for is the same mix-up in a different notation.

  5. 5

    Label every line as you write it

    Write dydx=\frac{dy}{dx}= or d2ydx2=\frac{d^{2}y}{dx^{2}}= before each result, not just the final number. It stops you mixing up which value belongs to which derivative, and it protects method marks even if a number later goes wrong.

One line to remember it

Differentiate once for the gradient, twice for its rate of change

dydx\frac{dy}{dx} is the slope; d2ydx2\frac{d^{2}y}{dx^{2}} is how the slope itself is changing. If a question says 'rate of change of the gradient', 'concavity' or 'acceleration', differentiate again, and remember a stationary point always gives dydx=0\frac{dy}{dx}=0, so only d2ydx2\frac{d^{2}y}{dx^{2}} can reveal its nature.

Practice where the error hides

Q1[3 marks]

The curve y=x33x29x+5y=x^{3}-3x^{2}-9x+5 has gradient function dydx\frac{dy}{dx}. Find (a) dydx\frac{dy}{dx}, and (b) d2ydx2\frac{d^{2}y}{dx^{2}}, then evaluate both at x=2x=2.

Show worked solution

Differentiate yy once to get the gradient function, then differentiate again to get the rate of change of the gradient.

dydx=3x26x9\frac{dy}{dx}=3x^{2}-6x-9
d2ydx2=6x6\frac{d^{2}y}{dx^{2}}=6x-6

Substitute x=2x=2 into each derivative separately, do not reuse the same working for both.

dydxx=2=3(2)26(2)9=12129=9\left.\frac{dy}{dx}\right|_{x=2}=3(2)^{2}-6(2)-9=12-12-9=-9
d2ydx2x=2=6(2)6=126=6\left.\frac{d^{2}y}{dx^{2}}\right|_{x=2}=6(2)-6=12-6=6

So at x=2x=2 the gradient is 9-9 (the curve is falling steeply) while the second derivative is +6+6 (the gradient is increasing, so the curve is concave up there), two different numbers answering two different questions.

Q2[4 marks]

Show that the curve y=x33x29x+5y=x^{3}-3x^{2}-9x+5 has stationary points at x=1x=-1 and x=3x=3, and determine the nature of each using the second derivative.

Show worked solution

A stationary point occurs where dydx=0\frac{dy}{dx}=0.

dydx=3x26x9=3(x3)(x+1)=0  x=3 or x=1\frac{dy}{dx}=3x^{2}-6x-9=3(x-3)(x+1)=0\ \Rightarrow\ x=3\ \text{or}\ x=-1

Substituting dydx\frac{dy}{dx} into the test here would give 00 at both points, because that is exactly the equation just solved, it can never distinguish a maximum from a minimum. Differentiate again and use d2ydx2\frac{d^{2}y}{dx^{2}} instead.

d2ydx2=6x6\frac{d^{2}y}{dx^{2}}=6x-6
d2ydx2x=1=6(1)6=12<0  maximum\left.\frac{d^{2}y}{dx^{2}}\right|_{x=-1}=6(-1)-6=-12<0\ \Rightarrow\ \text{maximum}
d2ydx2x=3=6(3)6=12>0  minimum\left.\frac{d^{2}y}{dx^{2}}\right|_{x=3}=6(3)-6=12>0\ \Rightarrow\ \text{minimum}

So x=1x=-1 gives a maximum and x=3x=3 gives a minimum, the classification comes from the second derivative, never from the first.

Q3[3 marks]

A particle moves in a straight line so that its displacement, ss metres, from a fixed point OO at time tt seconds is given by s=t36t2+9ts=t^{3}-6t^{2}+9t, for t0t\ge0. Find the acceleration of the particle when t=4t=4.

Show worked solution

Velocity is the first derivative of displacement, and acceleration is the second derivative, velocity differentiated again.

v=dsdt=3t212t+9v=\frac{ds}{dt}=3t^{2}-12t+9

At t=4t=4 this gives the velocity, not the acceleration:

v(4)=3(4)212(4)+9=4848+9=9 m s1v(4)=3(4)^{2}-12(4)+9=48-48+9=9\ \text{m s}^{-1}

The acceleration needs a second differentiation.

a=d2sdt2=dvdt=6t12a=\frac{d^{2}s}{dt^{2}}=\frac{dv}{dt}=6t-12
a(4)=6(4)12=12 m s2a(4)=6(4)-12=12\ \text{m s}^{-2}

The velocity at t=4t=4 is 99 m/s; the acceleration at t=4t=4 is 1212 m/s2^{-2}, reporting the 99 as the acceleration is the same first/second-derivative mix-up, only in a kinematics setting.

How one-to-one teaching helps

Our teachers make you say out loud which derivative a question wants before you touch your pen,'gradient' or 'rate of change of the gradient', dydx\frac{dy}{dx} or d2ydx2\frac{d^{2}y}{dx^{2}}, because naming it first stops the second differentiation from getting skipped under exam pressure. We drill the stationary-point habit until it is automatic: dydx=0\frac{dy}{dx}=0 finds the point, d2ydx2\frac{d^{2}y}{dx^{2}} alone classifies it, never the other way round.

In a one-to-one lesson we also walk through the kinematics wording, displacement, velocity, acceleration, so the same notation slip does not resurface in a different topic. Teachers at spmaddmath.com.my are experienced, and lessons are online and taught in English.

Message us on WhatsApp to arrange a one-hour paid class from RM50/hr at the teacher's rate.

Get 1-to-1 help.

Book a Trial Class

Frequently asked questions

What is the difference between dy/dx and d²y/dx²?

dydx\frac{dy}{dx} is the first derivative, the gradient of the curve at a point. d2ydx2\frac{d^{2}y}{dx^{2}} is the second derivative, dydx\frac{dy}{dx} differentiated again, which measures how fast the gradient itself is changing (the curve's concavity).

Why can't I use dy/dx in the second-derivative test?

Because a stationary point is defined by dydx=0\frac{dy}{dx}=0, substituting it into the test always gives 00 and can never tell you whether the point is a maximum or a minimum. You must differentiate again and substitute the xx-value into d2ydx2\frac{d^{2}y}{dx^{2}} instead.

In kinematics, which derivative gives acceleration?

Acceleration is the second derivative of displacement with respect to time, a=d2sdt2=dvdta=\frac{d^{2}s}{dt^{2}}=\frac{dv}{dt}, velocity differentiated again, not velocity itself.

Does a positive d²y/dx² always mean a maximum?

No, the opposite. A positive d2ydx2\frac{d^{2}y}{dx^{2}} at a stationary point means a minimum (concave up); a negative value means a maximum (concave down).

Mixing up that sign is a separate, common slip from confusing dy/dx with d²y/dx² itself.

Source:SRC-DSKP-EN

Written by the spmaddmath.com.my editorial team.· Last updated 5 September 2026

Ready to get started?

Book a Trial Classfrom RM50/hr · One-hour paid trial · Same-day reply
Book a Trial ClassOne-hour paid trial · Same-day reply