The remainder theorem says that dividing P(x) by x−a leaves remainder P(a). So what happens when P(a)=0? There’s no remainder, which means x−a divides evenly: it’s a factor. This is the factor theorem, and it’s how you factor cubics and quartics that don’t fit any of the familiar patterns.
Integer zeros. If P(x) has integer coefficients and x−b is a factor (with b an integer), then b is a factor of the constant term. For x3−2x2−5x+6, the only integers worth testing are the factors of 6: ±1,±2,±3,±6.
Why? If P(x)=(x−b)(…), then multiplying the constant terms of the factors gives the constant term of P(x). So b must divide it.
Rational zeros. If the leading coefficient isn’t 1, a zero might be a fraction. Any rational zero ab (in lowest terms) has b a factor of the constant term and a a factor of the leading coefficient. For 2x3+x2+5x−3, the candidates are
List the possible zeros and test them until you find a value a with P(a)=0.
Divide P(x) by x−a (synthetic division is quickest). The remainder should be 0.
Factor the quotient. If it’s a quadratic, factor it the usual way. If it’s a cubic, repeat steps 1 and 2.
Write P(x) as a product of all the factors, and check by expanding or by substituting a value.
Sometimes the last quadratic doesn’t factor. If its discriminant b2−4ac is negative, it has no real zeros, so it can’t be factored any further using real numbers.
Mixing up the zero and the factor. If P(−3)=0, the factor is x+3, not x−3. The factor is ”x minus the zero”.
Testing values that can’t work. For integer coefficients, only factors of the constant term can be integer zeros. Testing x=5 for x3−2x2−5x+6 is wasted effort, because 5 doesn’t divide 6.
Forgetting fractions when the leading coefficient isn’t 1. If no integer works for 2x3+x2+5x−3, try ±21 and ±23 before deciding it can’t be factored.
Stopping too early. After dividing out one factor, check whether the quotient factors further. (x−1)(x2−x−6) isn’t fully factored until you write (x−1)(x−3)(x+2).
Forcing a quadratic to factor. If the discriminant of the leftover quadratic is negative, it doesn’t factor over the real numbers. That’s a correct final answer, not a mistake.
1. (Warm-up) Let P(x)=x3−x2−4x−6. Is x−3 a factor? Is x+2 a factor?
Solution
P(3)=27−9−12−6=0, so x−3 is a factor.
P(−2)=−8−4+8−6=−10=0, so x+2 is not a factor.
2. (Warm-up) List the possible integer zeros of x3+2x2−5x−6.
Solution
The factors of the constant term, −6: ±1,±2,±3,±6.
3. (Core) Factor x3+2x2−5x−6 fully.
Solution
Try x=−1: P(−1)=−1+2+5−6=0, so x+1 is a factor.
−1112−11−5−1−6−660
x2+x−6=(x+3)(x−2), so
x3+2x2−5x−6=(x+1)(x+3)(x−2)
4. (Core) Factor x3−7x+6 fully.
Solution
P(1)=1−7+6=0, so x−1 is a factor. Use 0 for the missing x2 term:
111011−71−66−60
x2+x−6=(x+3)(x−2), so x3−7x+6=(x−1)(x−2)(x+3).
5. (Core) Find k so that x+2 is a factor of x3+kx2−2x+8. Then factor the polynomial as fully as possible.
Solution
x+2 is a factor when P(−2)=0:
−8+4k+4+8=0⇒4k=−4⇒k=−1
Now divide x3−x2−2x+8 by x+2:
−211−1−2−3−2648−80
x3−x2−2x+8=(x+2)(x2−3x+4). The discriminant of x2−3x+4 is 9−16=−7<0, so it doesn’t factor further.
6. (Core) For P(x)=3x3+4x2−5x−2:
(a) List the possible rational zeros.
(b) Factor P(x) fully.
Solution
(a) ab with b a factor of −2 and a a factor of 3: ±1,±2,±31,±32.
(b) P(1)=3+4−5−2=0, so x−1 is a factor.
133437−572−220
3x2+7x+2=(3x+1)(x+2), so P(x)=(x−1)(3x+1)(x+2).
7. (Core) Factor x4−2x3−7x2+8x+12 fully.
Solution
P(−1)=1+2−7−8+12=0, so x+1 is a factor.
−111−2−1−3−73−4841212−120
Factor the cubic x3−3x2−4x+12 by grouping: x2(x−3)−4(x−3)=(x−3)(x2−4)=(x−3)(x−2)(x+2).
x4−2x3−7x2+8x+12=(x+1)(x−3)(x−2)(x+2)
8. (Challenge) Find a and b so that both x−1 and x+2 are factors of x3+ax2+bx−6. Then find the third factor.
Solution
P(1)=0: 1+a+b−6=0, so a+b=5.
P(−2)=0: −8+4a−2b−6=0, so 4a−2b=14, or 2a−b=7.
Adding the two equations: 3a=12, so a=4 and b=1.
The polynomial is x3+4x2+x−6. Since (x−1)(x+2)=x2+x−2, the third factor must be linear with leading coefficient 1 and constant term −2−6=3. So it’s x+3.