Nature of roots

Two quadratic equations can look equally messy and yet behave completely differently once you try to solve them — one splits cleanly into two real answers, another gives a single repeated answer, and a third refuses to give any real answer at all. You do not need to solve any of them to know which is which. Buried inside the quadratic formula is a single expression, $b^2-4ac$ , called the discriminant, and its sign alone settles the question. This is the standard ICSE Class 10 tool for answering "how many real roots does this equation have" and for the reverse problem — given that an equation must have equal roots, or must have no real roots, find the unknown coefficient that makes that true. Both directions rest on exactly the same idea, so mastering the sign logic of $D=b^2-4ac$ once pays off on every variant the exam throws at it.

The core insight

The sign of the discriminant $D=b^2-4ac$ of $ax^2+bx+c=0$ tells you the nature of the roots — positive gives two distinct real roots, zero gives two equal real roots, negative gives no real roots — without ever solving the equation.

The core idea

Standard form first. Every quadratic equation must be arranged as $ax^2+bx+c=0$ with $a \ne 0$ before you read off $a$ , $b$ , $c$ . Skipping this step is the single most common source of sign errors, since a stray term left on the wrong side flips a coefficient's sign.

The discriminant. For the equation in standard form,

ax^2+bx+c=0, \quad a \ne 0,

define

D = b^2 - 4ac.

where: $a$ is the coefficient of $x^2$ , $b$ is the coefficient of $x$ , and $c$ is the constant term.

Why the sign of $D$ decides everything. The quadratic formula is

x = \dfrac{-b \pm \sqrt{D}}{2a}.

where: the $\pm$ produces two candidate values of $x$ , and $D$ sits directly under the square root, so it controls whether those two values are real and whether they are actually different from each other.

The three-way classification.

$D>0$ : two distinct real roots, since $\sqrt{D}$ is a positive real number and the $\pm$ genuinely splits into two different values.
$D=0$ : two equal real roots, since $\sqrt{D}=0$ makes the $\pm$ irrelevant and both roots collapse to $x=-\dfrac{b}{2a}$ .
$D<0$ : no real roots, since the square root of a negative number is not a real number, so no real $x$ satisfies the equation.

A geometric picture, in words. The graph of $y=ax^2+bx+c$ is a parabola, and the real roots of $ax^2+bx+c=0$ are exactly the x-intercepts, the points where that parabola meets the x-axis. Two distinct real roots mean the curve crosses the axis at two separate points; equal roots mean the curve only just touches the axis at its own turning point without crossing it; no real roots mean the whole parabola stays entirely above the axis (if $a>0$ ) or entirely below it (if $a<0$ ), never meeting it at all. This picture is why the exam phrase "nature of the roots" is really asking "how does this parabola sit relative to the x-axis," and it is a handy sanity check: a positive leading coefficient with a very negative constant term should usually give two crossings, which you can confirm once you compute $D$ .

Common exam framing. Most questions do not simply hand you $a$ , $b$ , $c$ as plain numbers. Instead they give you an equation containing an unknown letter, typically $k$ or $p$ , state a condition on the roots in words — "equal roots," "real and distinct roots," or "no real roots" — and ask you to find the unknown. The method never changes: identify $a$ , $b$ , $c$ in terms of the unknown, write down $D$ as an expression in that unknown, translate the stated condition into $D=0$ , $D>0$ , or $D<0$ , and solve the resulting equation or inequality.

Figure

Three parabolas showing a positive discriminant crossing the x axis twice a zero discriminant touching it once and a negative discriminant never meeting it

Why it works

Derivation AWhy the discriminant sign fixes the number of real roots

The goal here is to show, starting only from completing the square on $ax^2+bx+c=0$ , exactly why the expression $b^2-4ac$ is the one quantity that decides whether the quadratic formula's two candidate roots are real and distinct, real and equal, or not real at all, so that the classification rule can be trusted rather than merely memorised.

Step 1. Start from $ax^2+bx+c=0$ with $a \ne 0$ and divide throughout by $a$ : $x^2 + \dfrac{b}{a}x + \dfrac{c}{a} = 0$ .

Step 2. Complete the square on the left: $\left(x+\dfrac{b}{2a}\right)^2 = \dfrac{b^2}{4a^2} - \dfrac{c}{a} = \dfrac{b^2-4ac}{4a^2}$ . The right-hand side is exactly $D$ divided by the positive quantity $4a^2$ .

Step 3. Take the square root of both sides: $x + \dfrac{b}{2a} = \pm\dfrac{\sqrt{D}}{2a}$ , so $x = \dfrac{-b\pm\sqrt{D}}{2a}$ , which is the familiar quadratic formula rebuilt from scratch. Because $4a^2$ is always positive, the sign of the right-hand side under the square root in Step 2 is decided purely by the sign of $D$ .

Step 4. Read off the three cases directly from this square root. If $D>0$ , $\sqrt{D}$ is a positive real number, so the $+$ and $-$ choices genuinely differ and give two distinct real roots. If $D=0$ , $\sqrt{D}=0$ , so both choices collapse onto the single value $x=-\dfrac{b}{2a}$ — an equal (repeated) root. If $D<0$ , the quantity under the square root is negative, and no real number squares to a negative number, so there is no real value of $x$ that satisfies the original equation.

Worked example

Question. Find the value(s) of $k$ for which the equation $kx^2 - 4x + 1 = 0$ has equal roots, and verify the result.

Step 1. Compare with $ax^2+bx+c=0$ : here $a=k$ , $b=-4$ , $c=1$ , with $a\ne0$ requiring $k \ne 0$ .

Step 2. For equal roots the condition is $D=0$ : $D = b^2-4ac = (-4)^2 - 4(k)(1) = 16-4k$ .

Step 3. Set $D=0$ and solve: $16-4k=0 \Rightarrow 4k=16 \Rightarrow k=4$ . Since $k=4 \ne 0$ , the quadratic condition $a\ne0$ is satisfied, so $k=4$ is a valid answer.

Step 4. Verify by substitution: with $k=4$ the equation becomes $4x^2-4x+1=0$ , which factorises as $(2x-1)^2=0$ , giving the repeated root $x=\tfrac12,\tfrac12$ — confirming equal roots. Notice the left side is a perfect square, which is always true whenever $D=0$ , since the equation reduces to $a\left(x+\dfrac{b}{2a}\right)^2=0$ .

The required value is $k=4$ .

Competitive stretch

If instead the question asked for $kx^2-4x+1=0$ to have no real roots, the same setup gives $D=16-4k<0 \Rightarrow k>4$ (together with $k \ne 0$ , which is automatic once $k>4$ ), so every $k$ strictly greater than $4$ works — a useful pattern for "find the range of $k$ " variants.

Common traps

Forgetting $a \ne 0$ : after solving for the unknown, always check it does not make the coefficient of $x^2$ zero, or the equation stops being quadratic.
Sign slips inside $b^2-4ac$ , especially when $b$ or $c$ is negative — substitute in brackets, e.g. $(-4)^2$ , never $-4^2$ .
Treating "no real roots" as "no solution at all" — a negative discriminant only means the roots are not real numbers; at this level you simply write "no real roots."
Confusing "real and distinct" ( $D>0$ only) with "real" (which allows $D \ge 0$ , covering both distinct and equal cases) — read the exact wording of the question before choosing $>$ , $=$ , or $<$ .

Ask the tutor: Want to try a "find the range of k" no-real-roots problem, or see why an equal-roots quadratic always factors as a perfect square?

Why do we call D the discriminant instead of just calling it part of the formula

Because its only job is to discriminate, that is, distinguish, between the three possible outcomes for the roots before you do any further work. You could in principle always apply the full quadratic formula and see what happens, but computing just $b^2-4ac$ is faster and is exactly what most exam questions ask for — "find the nature of the roots," not "solve the equation."

If D is a perfect square like 25 does that change anything about the classification

No — the classification only cares about the sign of $D$ , so $D=25>0$ still just means two distinct real roots. A perfect square $D$ is a bonus: it tells you the roots will also be rational numbers, since $\sqrt{D}$ comes out as a whole number in the quadratic formula. This is worth noticing but is not part of the required classification.

Can a and c have the same sign and still give D greater than 0

Yes, easily — the sign of $D=b^2-4ac$ depends on the size of $b^2$ relative to $4ac$ , not on the individual signs of $a$ and $c$ . For example $a=1,b=10,c=1$ gives $D=100-4=96>0$ even though $a$ and $c$ are both positive. Always compute $D$ directly rather than trying to guess its sign from $a$ and $c$ alone.