01Bounds and accuracy · E1.10
A measurement a rounded to the nearest step h represents a − h/2 ≤ true value < a + h/2. For a product of positive quantities, multiply both lower bounds or both upper bounds. For a positive quotient, use lower numerator / upper denominator for the lower bound and upper numerator / lower denominator for the upper bound. Check signs and units first; retain intermediate precision.
02Surds and rationalisation · E1.18
For non-negative a and b, √(ab) = √a × √b. Extract perfect-square factors and combine only like surds. In general √(a + b) is not √a + √b. For a/(√b + c), with b ≥ 0, the original denominator must be non-zero. Multiply by conjugate √b − c only if it too is non-zero, equivalently b ≠ c². If the conjugate is zero, that step is invalid; simplify the original denominator directly.
03Standard form · E1.8
A positive number in standard form is a × 10ⁿ with 1 ≤ a < 10 and integer n. Multiply coefficients and add powers for products; divide coefficients and subtract powers for quotients. For addition, first express both terms using the same power of 10. Finally renormalise the coefficient into [1, 10) and check the order of magnitude.
04Compound percentages · E1.13 / E1.17
An increase of r% uses multiplier 1 + r/100; a decrease uses 1 − r/100. For n periods at a constant rate, final value = initial value × multiplierⁿ. Different rates require separate multipliers. Reverse the change by dividing by their product. An equal percentage increase and decrease do not cancel because their bases differ.
05Direct and inverse proportion · E2.8
Translate y ∝ xᵖ into y = kxᵖ and y ∝ 1/xᵖ into y = k/xᵖ. The power may describe a square, cube or root; do not assume every relation is y = kx. Find k from a known pair, then substitute the new value. A denominator must be non-zero. A real √x requires x ≥ 0, and 1/√x requires x > 0. Check that the direction of change is sensible.
06Quadratic equations · E2.2 / E2.5
Rearrange to ax² + bx + c = 0 with a ≠ 0. Try factorisation or completing the square; in general x = (−b ± √(b² − 4ac))/(2a). The discriminant gives two, one repeated, or no real roots. The form a(x − h)² + k also identifies the vertex (h, k). Retain exact surds when requested and check by substitution.
07Algebraic fractions · E2.3 / E2.5
Record values that make any original denominator zero before simplifying. Factorise and cancel common factors, never individual terms in a sum. Use a common denominator for addition. In a fractional equation, multiply through by the common denominator while retaining the original restrictions; check every candidate solution.
08Simultaneous equations · E2.5
For two linear equations, use elimination or substitution. When one equation is quadratic, isolate a variable in the linear equation and substitute into the other. There may be two solution pairs. Find the matching second coordinate for each root and check each pair in both original equations.
09Inequalities · E2.6
Add or subtract the same quantity without changing direction. Multiplying or dividing by a negative reverses the inequality. Apply operations to all three parts of a compound inequality. On a number line, use an open point for < or > and a closed point for ≤ or ≥. For two variables, use a dashed boundary for a strict inequality and a solid boundary when equality is allowed; a test point determines the correct side.
10Functions and inverses · E2.13
A function f assigns one output f(x) to each allowed input x. The domain contains allowed inputs; the range contains attainable outputs. To find an inverse, write y = f(x), solve for x, then rename variables. The function must be one-to-one on its chosen domain: for x², for example, restrict x ≥ 0 to obtain inverse √x. The notation f⁻¹ does not mean 1/f. In gf(x), apply f first, then g.
11Sequences and the nth term · E2.7
For an arithmetic sequence, Tₙ = a + (n − 1)d with first term a and common difference d. Constant second differences suggest a quadratic rule an² + bn + c, with second difference 2a; subtract an² and analyse the remainder. A constant ratio suggests geometric rule arⁿ⁻¹. Cubic rules or simple combinations can also occur. A finite list does not uniquely determine every possible sequence: check the specified type of rule.