The Student Room Group

Factorising quadratics when a isn't 1

I got taught then when factorising a quadratic when a isn't 1 to just basically do trial and error and I wanted to know if there was a more efficient method than just trying a bunch of combinations.
Original post by RandomPancake
I got taught then when factorising a quadratic when a isn't 1 to just basically do trial and error and I wanted to know if there was a more efficient method than just trying a bunch of combinations.

What do you mean by trial and error?

Assuming roots of a quadratic are rational (meaning your leading coefficient is tyically not 1) then you can factorise it.

Let's say the roots are x=pqx = -\dfrac{p}{q} and x=rsx = - \dfrac{r}{s} and these correspond to linear factors (qx+p)(qx+p) and (sx+r)(sx+r) being zero.

Therefore, the two roots satisfy the quadratic equation (qx+p)(sx+r)=0(qx+p)(sx+r)=0, i.e. qsx2+(qr+ps)x+pr=0qs x^2 + (qr+ps) x + pr = 0

The leading coefficient is qsqs.

Here qr,psqr,ps are two magic numbers which add to give the middle coefficient, but they also multiply to give qrpsqrps which is exactly the same as the leading coefficient qsqs multiplying the constant term prpr.

All this is to say that there some logic involved here in factorising. When trying to factorise ax2+bx+c=0ax^2 + bx + c = 0 you should be splitting your bb into two magic numbers which add to bb yet multiply to acac.

As they multiply to acac, you just need to look at the factor pairs of this product and see what adds to bb.

Example: Factorise 2x2+5x+22x^2 + 5x + 2.

Look for two numbers which add to 55 and multiply to 2×2=42\times 2 = 4.

These are 1,41,4. Then rewrite your quadratic as (2x2+x)+(4x+2)(2x^2 + x) + (4x+2) [we have split 5x=x+4x5x=x+4x] and factorise fully each bracket to get

x(2x+1)+2(2x+1)x(2x+1) + 2(2x+1)

Finally note that (2x+1)(2x+1) is a bracket occuring twice in this expression therefore it can be factored out overall to give

(2x+1)(x+2)(2x+1)(x+2)

This is the factorisation. No trial and error needed. Just find the two magic numbers by looking at factor pairs of acac which add to bb.
(edited 5 months ago)
Reply 2
Original post by RDKGames
What do you mean by trial and error?

Assuming roots of a quadratic are rational (meaning your leading coefficient is tyically not 1) then you can factorise it.

Let's say the roots are x=pqx = -\dfrac{p}{q} and x=rsx = - \dfrac{r}{s} and these correspond to linear factors (qx+p)(qx+p) and (sx+r)(sx+r) being zero.

Therefore, the two roots satisfy the quadratic equation (qx+p)(sx+r)=0(qx+p)(sx+r)=0, i.e. qsx2+(qr+ps)x+pr=0qs x^2 + (qr+ps) x + pr = 0

The leading coefficient is qsqs.

Here qr,psqr,ps are two magic numbers which add to give the middle coefficient, but they also multiply to give qrpsqrps which is exactly the same as the leading coefficient qsqs multiplying the constant term prpr.

All this is to say that there some logic involved here in factorising. When trying to factorise ax2+bx+c=0ax^2 + bx + c = 0 you should be splitting your bb into two magic numbers which add to bb yet multiply to acac.

As they multiply to acac, you just need to look at the factor pairs of this product and see what adds to bb.

Example: Factorise 2x2+5x+22x^2 + 5x + 2.

Look for two numbers which add to 55 and multiply to 2×2=42\times 2 = 4.

These are 1,41,4. Then rewrite your quadratic as (2x2+x)+(4x+2)(2x^2 + x) + (4x+2) [we have split 5x=x+4x5x=x+4x] and factorise fully each bracket to get

x(2x+1)+2(2x+1)x(2x+1) + 2(2x+1)

Finally note that (2x+1)(2x+1) is a bracket occuring twice in this expression therefore it can be factored out overall to give

(2x+1)(x+2)(2x+1)(x+2)

This is the factorisation. No trial and error needed. Just find the two magic numbers by looking at factor pairs of acac which add to bb.

Thanks for the explanation!
We got taught to find factors of a and put them into the brackets first and then fill in the rest of the bracket with factors of c and do trial and error until the x values add up to b.
Original post by RandomPancake
Thanks for the explanation!
We got taught to find factors of a and put them into the brackets first and then fill in the rest of the bracket with factors of c and do trial and error until the x values add up to b.

OK, I see.

That works too, but this above method you might prefer it. Significantly fewer options to check when looking at factors of ac in one go rather than choosing a factor pair of a then looking at factors of c. This is great because it always works, although you might find the trial and error to be quicker if there are not many factors to consider.

Try both methods on this one example:

12x2+7x+1212x^2 + 7x + 12

to get (3x+4)(4x3)(3x+4)(4x-3).


See this video to help you with this method (called method 2 in the video):

https://www.youtube.com/watch?v=cbbEFYpeWV8
(edited 5 months ago)
Reply 4
Original post by RDKGames
OK, I see.

That works too, but this the above method you might prefer it. Significantly fewer options to check when looking at factors of ac in one go rather than choosing a factor pair of a then looking at factors of c.

Try both methods on this one example:

12x2+7x+1212x^2 + 7x + 12

to get (3x+4)(4x3)(3x+4)(4x-3).

The method you suggested definitely makes it a lot less time consuming and also makes cases where a and c have a ridiculous amount of factors (eg 12) a lot easier to work with. Using the method I got taught to factorise it took a lot longer than the method you showed and also took a lot more working out. Thanks once again!
Reply 5
You can actually reduce any factorisation problem (quadratic) to factorising a quadratic with leading term 1 - meaning the coefficient of x2 x^2 is 1.
Say you have ax2+bx+c ax^2+bx+c . Then multiply by a a which gives a2x2+abx+ac a^2x^2+abx+ac which is also (ax)2+b(ax)+ac (ax)^2+b(ax)+ac .
Now substitute y=ax y=ax which gives us y2+by+ac y^2+by+ac . This has coefficient 1 and so just factorise it as you know how. For the final step just divide by an and this will cancel with the factors in some way.

EXAMPLE
2x23x2 2x^2-3x-2 .
The value of a a is 2 so we multiply by 2 giving 4x26x4 4x^2-6x-4 . Now we set y=2x y=2x and we have y23y4 y^2-3y-4 .
You factorise this easily as (y4)(y+1)=(2x4)(2x+1) (y-4)(y+1)=(2x-4)(2x+1) .
Therefore we have 4x26x4=(2x4)(2x+1) 4x^2-6x-4 =(2x-4)(2x+1) but we can’t to factorise half of this, namely 2x23x2 2x^2-3x-2 and so we just divide by 2 giving
2x23x2=(x2)(2x+1) 2x^2-3x-2=(x-2)(2x+1) .

This method is not necessarily the most efficient, but all I’m showing is that actually it always can be boiled down to something you can easily factorise.

Quick Reply

Latest