HookThe spacecraft that burned up because of one formula
On 23 September 1999, NASA lost the Mars Climate Orbiter — a probe that cost about 327 million US dollars — as it slipped behind the planet and was never heard from again. The inquiry found no broken part and no software crash. The fault was a formula fed the wrong numbers. Lockheed Martin's ground software reported the thrust from a manoeuvre in pound-force-seconds; NASA's navigation code substituted those same digits into its equations as if they were newton-seconds, a unit about 4.45 times larger. The orbiter aimed for a safe altitude of roughly \(226\) km above Mars but arrived at about \(57\) km, deep in the atmosphere, and burned up. Same formula, same digits, catastrophically wrong values.
Everything in A2.1 is the discipline that would have caught that error: handling algebra so carefully that a wrong sign, a mis-substituted value or a dropped unit never survives to the next line. You will learn the notation that lets \(ab\) stand for \(a\times b\), the exact vocabulary that separates an equation from an identity, how to substitute numbers into scientific formulae without the NASA slip, the manipulation rules — indices, expanding brackets, factorising — that compress algebra, how to rearrange a formula to make any letter its subject, how to write a rigorous proof, and how to treat an expression as a function you can run forwards, backwards and in sequence. Get these automatic and the rest of Higher algebra is bookkeeping; fumble them and, like NASA, every later line inherits the mistake.
ModelThe grammar of algebra — notation and the words the mark scheme uses
Algebra drops the multiplication sign to save ink and confusion. \(ab\) means \(a\times b\); \(3y\) means \(3\times y\); \(a^2\) means \(a\times a\) and \(a^3\) means \(a\times a\times a\); a division \(a\div b\) is written as the fraction \(\dfrac{a}{b}\); and a fractional coefficient like \(\tfrac{2}{3}x\) is just two-thirds of \(x\). Brackets group whatever is inside so it is treated as a single object: \(2(x+5)\) is twice the whole of \(x+5\), which is why it expands to \(2x+10\) and not \(2x+5\).
Higher papers also test the words, because the command word tells you what is legal. A term is a single product of numbers and letters, like \(4x^2\). An expression is terms added or subtracted with no equals sign, like \(4x^2-3x+1\) — you can simplify it but never 'solve' it. An equation has an equals sign and is true only for particular values, so it can be solved. A formula is a relationship between variables that stays true whatever you put in, like \(A=\pi r^2\). An identity is true for every value and is written with the three-bar sign \(\equiv\), as in \((x+1)^2\equiv x^2+2x+1\). An inequality uses \(\lt,\ \gt,\ \leq,\ \geq\), and a factor is something that divides exactly into an expression. Naming these correctly is worth marks in its own right on the 'is this an equation or an identity?' style questions.
MechanismSubstitution — feeding numbers into a formula without the NASA slip
To substitute, replace every letter with its value inside brackets, then evaluate using the priority of operations (brackets, then powers and roots, then \(\times\) and \(\div\), then \(+\) and \(-\)). The brackets are not optional decoration: they are the single habit that stops the most common Higher error, mishandling negatives. Because a power binds tighter than a subtraction sign, \(-3^2\) means \(-(3^2)=-9\), whereas \((-3)^2=9\). Write \((-3)\) every time and the ambiguity disappears.
Scientific formulae are examined directly — the kinematics ('SUVAT') equations, \(E_k=\tfrac12 mv^2\), Ohm's law and the like. The method is identical: substitute in brackets, respect the units, and compute in order. A value in the wrong unit is exactly the Mars Orbiter failure at GCSE scale, so if a question gives a time in minutes and a speed per hour, convert before you substitute, not after.
A body starts at \(u=3\) m/s and accelerates at \(a=-2\) m/s\(^2\) for \(t=4\) s. Find its displacement using \(s=ut+\tfrac12 a t^2\).
Substitute in brackets: \(s=(3)(4)+\tfrac12(-2)(4)^2\). Powers first: \((4)^2=16\). Then the products: \(s=12+\tfrac12(-2)(16)=12+(-16)\). Finally the addition: \(s=-4\) m. The negative answer is meaningful — the deceleration was strong enough that the body has ended up \(4\) m behind where it started. A student who wrote \(-2\times 4^2\) as \(-2\times 4 \times 4\) without the bracket often loses the sign and reports \(+16\) instead of \(-16\); the bracket around \((-2)\) and around \((4)^2\) is what protects the mark.
MechanismSimplifying — like terms, indices, expanding and factorising
Four manipulation skills sit under A4 and Higher tests them together. First, collect like terms: only terms with the identical letter-and-power part combine, so \(2x+3x=5x\) but \(2x+3x^2\) will not simplify because \(x\) and \(x^2\) are different objects. Second, the laws of indices: \(a^m\times a^n=a^{m+n}\), \(a^m\div a^n=a^{m-n}\), \((a^m)^n=a^{mn}\), \(a^0=1\), \(a^{-n}=\dfrac{1}{a^n}\) and \(a^{1/2}=\sqrt{a}\). These let you rewrite \(\dfrac{6x^5}{2x^2}\) as \(3x^3\) in one line.
Third, expanding brackets. Two binomials multiply term-by-term: \((x+3)(x-2)=x^2-2x+3x-6=x^2+x-6\). Three brackets are expanded two at a time. Fourth, factorising reverses this. A simple quadratic \(x^2+5x+6\) factorises to \((x+2)(x+3)\) because \(2\) and \(3\) multiply to \(6\) and add to \(5\). A quadratic with a leading coefficient, \(ax^2+bx+c\), needs the 'splitting the middle term' method. The difference of two squares is the pattern worth memorising: \(a^2-b^2=(a-b)(a+b)\), so \(x^2-9=(x-3)(x+3)\). Finally, algebraic fractions are simplified by factorising top and bottom and cancelling any common bracket — never by cancelling individual terms.
Factorise \(6x^2+11x+3\), then simplify \(\dfrac{6x^2+11x+3}{2x+3}\).
For \(ax^2+bx+c\) with \(a=6,\ b=11,\ c=3\), find two numbers that multiply to \(ac=18\) and add to \(11\): those are \(9\) and \(2\). Split the middle term: \(6x^2+9x+2x+3\). Factorise in pairs: \(3x(2x+3)+1(2x+3)\). The bracket \((2x+3)\) is now common, so \(6x^2+11x+3=(3x+1)(2x+3)\).
Now the fraction: \(\dfrac{(3x+1)(2x+3)}{2x+3}\). The \((2x+3)\) cancels top and bottom, leaving \(3x+1\). Notice you could only cancel because the numerator was fully factorised first — cancelling the \(3\)s or the \(x\)s out of the original un-factorised fraction would have been wrong.
MechanismChanging the subject — even when the subject appears twice
Rearranging a formula means applying inverse operations to both sides until the letter you want stands alone. Adding undoes subtracting, multiplying undoes dividing, and squaring undoes a square root. The order is the reverse of the order the operations were applied to your target letter — you peel the onion from the outside in.
The Higher twist is the case where the subject appears twice. You cannot isolate a letter that is in two places at once, so the fix is a three-step ritual: gather every term containing the subject onto one side, factorise the subject out as a common factor, then divide by the bracket that is left. This is exactly the manoeuvre that turns a formula 'stuck' with the variable on both sides into a clean single statement — and it is one of the most reliably examined 4-mark questions on the whole paper.
Make \(x\) the subject of \(y=\dfrac{x+2}{x-3}\).
Multiply both sides by \((x-3)\) to clear the fraction: \(y(x-3)=x+2\). Expand the left: \(yx-3y=x+2\). Now \(x\) appears twice, so gather the \(x\)-terms on one side and everything else on the other: \(yx-x=2+3y\). Factorise \(x\) out of the left: \(x(y-1)=2+3y\). Finally divide by the bracket: \(x=\dfrac{2+3y}{y-1}\). The pivotal step is the factorisation \(x(y-1)\); without it there is no way to divide the \(x\) out, because it is tangled in two terms.
ModelEquation or identity — and how to write a proof that earns full marks
The difference examiners hammer is this: an equation is true for particular values of the unknown, so \(2x+1=7\) is only true when \(x=3\). An identity is true for every value, so \((x+1)^2\equiv x^2+2x+1\) holds whatever \(x\) is, which is why it earns the three-bar \(\equiv\). To prove an identity you take one side and manipulate it until it becomes the other; you never move terms across the \(\equiv\) as if solving.
Rigorous algebraic proof — the 'show that…' and 'prove that…' questions worth up to four marks — rests on representing numbers generally. An even number is \(2n\); an odd number is \(2n+1\); consecutive integers are \(n\) and \(n+1\); a multiple of \(3\) is \(3n\). You then manipulate the general expression and read a conclusion from its final factored form. The single biggest error is 'proving' a statement by trying one example — testing \(n=4\) shows a case, never the general rule, and scores nothing.
Prove that the sum of the squares of two consecutive integers is always odd.
Let the two consecutive integers be \(n\) and \(n+1\), where \(n\) is any integer. Their squares sum to \(n^2+(n+1)^2\). Expand the second square: \(n^2+(n^2+2n+1)=2n^2+2n+1\). Now factor a \(2\) out of the first two terms: \(2(n^2+n)+1\). Because \(2(n^2+n)\) is a multiple of \(2\) it is even, and one more than an even number is always odd. Since \(n\) stood for any integer, the result holds for every pair of consecutive integers, not just a lucky example — which is what turns a demonstration into a proof.
CaseFunctions — running an expression forwards, backwards and in sequence
A function is a machine: an input \(x\) goes in, a rule acts on it, an output comes out. The notation \(f(x)=2x+1\) names that machine \(f\), so \(f(3)\) means substitute \(3\) to get \(7\). The Higher extensions are the inverse and the composite.
The inverse function \(f^{-1}\) runs the machine backwards, undoing whatever \(f\) did. To find it, write \(y=f(x)\), rearrange to make \(x\) the subject, then swap the letters back. A composite function chains two machines: \(fg(x)\) means do \(g\) first and feed its output into \(f\). The order is the trap — the function nearest the \(x\) acts first, and \(fg(x)\) is generally not the same as \(gf(x)\). Reading composites right-to-left, like the layers of an onion, keeps you out of trouble.
Let \(f(x)=2x+1\) and \(g(x)=x^2\).
Compute \(fg(3)\): the inner function \(g\) acts first, \(g(3)=3^2=9\), then \(f(9)=2(9)+1=19\). Now compute \(gf(3)\): here \(f\) acts first, \(f(3)=2(3)+1=7\), then \(g(7)=7^2=49\). The answers \(19\) and \(49\) differ, proving \(fg\neq gf\) — order is everything.
Find the inverse of \(f\): set \(y=2x+1\), rearrange for \(x\) as \(x=\dfrac{y-1}{2}\), then rename to get \(f^{-1}(x)=\dfrac{x-1}{2}\). Check it undoes \(f\): \(f(3)=7\) and \(f^{-1}(7)=\dfrac{7-1}{2}=3\), back where we started.
VocabularyKey terms the mark scheme pays for
TrapsMisconceptions that cost marks
ExamWhat examiners want
This section is examined on both the non-calculator (Paper 1) and calculator (Papers 2 and 3) papers, and almost every question is marked method-first, so show the intermediate lines. On substitution questions, write the formula, then the substituted line with every value in brackets, then the answer with its unit — a bare number in the wrong unit is the Mars Orbiter error and examiners deduct for it. On 'change the subject' questions where the variable appears twice, the two method marks live in the two structural steps: gathering the subject's terms on one side, and factorising the subject out; write those lines explicitly even if the algebra feels obvious.
For factorising \(ax^2+bx+c\), state the two numbers that multiply to \(ac\) and add to \(b\) before you split the middle term, so the examiner can follow the choice. On 'prove that' or 'show that' questions, open with a definition line ('Let the integer be \(n\)…'), never a worked example, and finish by pointing at the factored form that carries the conclusion (a factor of \(2\) for even, a \(+1\) for odd). With functions, write the inner evaluation of a composite first and label which function you applied — muddling \(fg\) with \(gf\) is the classic dropped mark, and a single check that your inverse returns the original input costs one line and saves the whole answer.