AQA A-Level Further Maths: Matrices — In-Depth Revision Guide (7367)

Matrices are one of the most important and most useful topics in AQA A-Level Further Mathematics (7367). A matrix is a compact way of packaging a whole system of numbers, and once you learn the rules of matrix algebra you can rotate and reflect the plane, solve simultaneous equations in three unknowns, and uncover the hidden directions along which a transformation simply stretches. The topic also reaches far beyond the exam: matrices are the language of computer graphics, machine learning, quantum mechanics, structural engineering and economics, so the time you invest here pays dividends in almost any STEM degree.

This guide is the in-depth companion for AQA's matrices course. It runs from matrix operations and multiplication, through determinants and inverses, into solving systems of equations, the geometry of linear transformations in 2D and 3D, and then the deeper structural ideas — eigenvalues and eigenvectors, diagonalisation, and the Cayley-Hamilton theorem. Each section gives the core skills, a worked example, the common pitfalls, and a link to the matching lesson.

The aim is not to replace working through problems. Matrix algebra becomes fast only once you have done enough multiplications, determinants and inverses that the patterns are automatic. The aim of this guide is to give you a clear map of what AQA can examine, in roughly the order they teach it, so your revision is targeted rather than scattered.

What the AQA 7367 Specification Covers

AQA A-Level Further Mathematics (7367) is a linear qualification assessed through three written papers, each 2 hours long, each worth 100 marks, and each carrying 33⅓% of the marks. Papers 1 and 2 are compulsory and cover the Pure content — which is where matrices, complex numbers and further calculus all sit. Paper 3 is the optional applications paper, on which you choose two of Mechanics, Statistics and Discrete Mathematics. Matrices are compulsory Pure, so they can appear on either Paper 1 or Paper 2, frequently as a long structured question that runs from a transformation through to eigenvectors.

Further Mathematics is one of the most highly valued A-Levels for STEM study, and is effectively expected for mathematics and many engineering and physics courses at Oxbridge and other competitive universities. The matrix and transformation content also rewards exactly the kind of careful, structured reasoning that the STEP and MAT admissions tests look for.

The table below shows the matrices sub-topics, what each involves, and why it matters across the course.

Topic	What it involves	Why it matters
Matrix operations	Addition, scalar multiplication, the zero and identity matrices	The basic algebra of matrices
Matrix multiplication	Row-by-column products, non-commutativity	The operation that does all the work
Determinants (2×2 and 3×3)	A single number measuring area/volume scaling	Tests invertibility; geometry of transformations
Inverse matrices	Undoing a transformation; $A^{-1}$A−1	Solving equations; reversing maps
Solving systems of equations	Writing $A\mathbf{x} = \mathbf{b}$Ax=b and inverting	Three-variable simultaneous equations
Linear transformations in 2D	Rotations, reflections, enlargements, shears	The geometric meaning of $2\times 2$2×2 matrices
Linear transformations in 3D	Rotations and reflections in space	Extends the geometry to three dimensions
Eigenvalues and eigenvectors	Directions a matrix only scales	The deep structure of a transformation
Diagonalisation	Writing $A = PDP^{-1}$A=PDP−1	Computing high powers; decoupling systems
The Cayley-Hamilton theorem	A matrix satisfies its own characteristic equation	Inverses and powers without extra work

These topics build on one another closely. Multiplication underpins everything; determinants decide invertibility; inverses solve systems; transformations give the geometry; and eigenvalues, diagonalisation and Cayley-Hamilton are the structural payoff. Revising them in order pays off.

Matrix Operations

A matrix is a rectangular array of numbers, described by its order $m \times n$m×n — $m$m rows by $n$n columns. The individual numbers are its entries (or elements). The basic operations are gentle: you add two matrices of the same order by adding corresponding entries, and you multiply by a scalar by multiplying every entry by that number. Matrices of different orders cannot be added at all.

Two special matrices recur constantly. The zero matrix has every entry zero and behaves like 0 in ordinary arithmetic. The identity matrix $I$I is square, with 1s down the leading diagonal and 0s elsewhere, and behaves like 1: multiplying any matrix by $I$I (of the right order) leaves it unchanged. Holding these two roles in mind makes matrix algebra feel much more like ordinary algebra.

A short worked example. Given $A = \begin{pmatrix} 1 & 2 \\ 3 & 0 \end{pmatrix}$A=(13​20​) and $B = \begin{pmatrix} 4 & -1 \\ 2 & 5 \end{pmatrix}$B=(42​−15​), find $2A - B$2A−B. Scale $A$A first: $2A = \begin{pmatrix} 2 & 4 \\ 6 & 0 \end{pmatrix}$2A=(26​40​). Subtract entry by entry: $2A - B = \begin{pmatrix} 2-4 & 4-(-1) \\ 6-2 & 0-5 \end{pmatrix} = \begin{pmatrix} -2 & 5 \\ 4 & -5 \end{pmatrix}$2A−B=(2−46−2​4−(−1)0−5​)=(−24​5−5​).

A common pitfall is trying to add matrices of different orders, which is undefined. Another is forgetting to apply the scalar to every entry, including the zeros and the diagonal. For practice across the basic operations, see the Introduction to Matrices lesson.

Matrix Multiplication

Matrix multiplication is the operation that gives matrices their power, and it is the one most worth drilling until it is automatic. To multiply $AB$AB, you take the dot product of each row of $A$A with each column of $B$B: the entry in row $i$i, column $j$j of the product is the sum of products of the entries of row $i$i of $A$A with column $j$j of $B$B. For this to work, the number of columns of $A$A must equal the number of rows of $B$B — an $m \times n$m×n matrix times an $n \times p$n×p matrix gives an $m \times p$m×p result.

The single most important fact, and the one AQA tests repeatedly, is that matrix multiplication is not commutative: in general $AB \neq BA$AB=BA. Sometimes $BA$BA is not even defined when $AB$AB is. Geometrically this is obvious — rotating then reflecting is not the same as reflecting then rotating. Order matters, always.

A short worked example. With $A = \begin{pmatrix} 1 & 2 \\ 3 & 4 \end{pmatrix}$A=(13​24​) and $B = \begin{pmatrix} 0 & 1 \\ 5 & 2 \end{pmatrix}$B=(05​12​), compute $AB$AB. The top-left entry is $(1)(0) + (2)(5) = 10$(1)(0)+(2)(5)=10; the top-right is $(1)(1) + (2)(2) = 5$(1)(1)+(2)(2)=5; the bottom-left is $(3)(0) + (4)(5) = 20$(3)(0)+(4)(5)=20; the bottom-right is $(3)(1) + (4)(2) = 11$(3)(1)+(4)(2)=11. So $AB = \begin{pmatrix} 10 & 5 \\ 20 & 11 \end{pmatrix}$AB=(1020​511​). Computing $BA$BA gives a different matrix — try it and confirm.

A common pitfall is multiplying entrywise like addition; multiplication is row-by-column, not element-by-element. Another is reversing the order out of habit. For worked products of various orders, see the Matrix Multiplication lesson.

Determinants (2×2 and 3×3)

The determinant of a square matrix is a single number that captures whether the matrix is invertible and, geometrically, the factor by which the transformation scales area (in 2D) or volume (in 3D). For a $2 \times 2$2×2 matrix,

$\det\!\begin{pmatrix} a & b \\ c & d \end{pmatrix} = ad - bc.$det(ac​bd​)=ad−bc.

For a $3 \times 3$3×3 matrix you expand along a row or column using cofactors (the "minors with alternating signs" method). The leading sign pattern across the top row is $+, -, +$+,−,+, and each term is the entry times the determinant of the $2\times 2$2×2 matrix left when its row and column are deleted.

The most important property is the link to invertibility: a matrix is invertible if and only if its determinant is non-zero. A matrix with determinant zero is called singular and cannot be inverted; geometrically it collapses the plane (or space) onto a line or point, so the transformation cannot be undone. A useful multiplicative property is $\det(AB) = \det(A)\det(B)$det(AB)=det(A)det(B).

A worked example. Find the determinant of $\begin{pmatrix} 2 & 1 & 0 \\ 1 & 3 & 1 \\ 0 & 2 & 4 \end{pmatrix}$​210​132​014​​ by expanding along the top row. This is $2\det\!\begin{pmatrix} 3 & 1 \\ 2 & 4 \end{pmatrix} - 1\det\!\begin{pmatrix} 1 & 1 \\ 0 & 4 \end{pmatrix} + 0 = 2(12 - 2) - 1(4 - 0) + 0 = 20 - 4 = 16$2det(32​14​)−1det(10​14​)+0=2(12−2)−1(4−0)+0=20−4=16. Since this is non-zero, the matrix is invertible.

A common pitfall is forgetting the alternating sign pattern in a $3\times 3$3×3 cofactor expansion. Another is computing $bc - ad$bc−ad instead of $ad - bc$ad−bc for the $2\times 2$2×2 case. For both sizes worked carefully, see the Determinants lesson.

Inverse Matrices

The inverse of a square matrix $A$A is the matrix $A^{-1}$A−1 such that $AA^{-1} = A^{-1}A = I$AA−1=A−1A=I. It exists exactly when $\det A \neq 0$detA=0. For a $2 \times 2$2×2 matrix there is a clean formula:

$\begin{pmatrix} a & b \\ c & d \end{pmatrix}^{-1} = \frac{1}{ad - bc}\begin{pmatrix} d & -b \\ -c & a \end{pmatrix}.$(ac​bd​)−1=ad−bc1​(d−c​−ba​).

In words: swap the leading-diagonal entries, negate the other two, and divide by the determinant. For a $3 \times 3$3×3 matrix the inverse is found via the matrix of cofactors: form the cofactor matrix, transpose it to get the adjugate, and divide by the determinant. AQA expects you to handle the $3\times 3$3×3 inverse, and it is mostly careful bookkeeping.

A vital property for solving equations is that the inverse of a product reverses the order: $(AB)^{-1} = B^{-1}A^{-1}$(AB)−1=B−1A−1. This mirrors getting dressed and undressed — to undo "socks then shoes" you take off the shoes first.

A worked example. Find the inverse of $A = \begin{pmatrix} 4 & 3 \\ 2 & 1 \end{pmatrix}$A=(42​31​). The determinant is $(4)(1) - (3)(2) = 4 - 6 = -2$(4)(1)−(3)(2)=4−6=−2. Applying the formula: $A^{-1} = \dfrac{1}{-2}\begin{pmatrix} 1 & -3 \\ -2 & 4 \end{pmatrix} = \begin{pmatrix} -\tfrac{1}{2} & \tfrac{3}{2} \\ 1 & -2 \end{pmatrix}$A−1=−21​(1−2​−34​)=(−21​1​23​−2​). Multiply back to check it gives $I$I.

A common pitfall is forgetting to divide by the determinant, or dividing only some entries. Another is reversing the swap-and-negate pattern. For the $3\times 3$3×3 adjugate method step by step, see the Inverse Matrices lesson.

Solving Systems of Equations with Matrices

A system of simultaneous equations can be written compactly as a single matrix equation $A\mathbf{x} = \mathbf{b}$Ax=b, where $A$A holds the coefficients, $\mathbf{x}$x is the column of unknowns, and $\mathbf{b}$b is the column of constants. If $A$A is invertible, the solution is simply $\mathbf{x} = A^{-1}\mathbf{b}$x=A−1b. This is the matrix method's headline application and a reliable route through three-variable systems that would be fiddly by elimination.

The determinant also tells you about the nature of the solution. If $\det A \neq 0$detA=0, there is a unique solution. If $\det A = 0$detA=0, the matrix is singular and the system has either no solutions (the equations are inconsistent — the planes have no common point) or infinitely many (the equations are dependent — the planes meet in a common line). Recognising the singular case is part of the question.

A worked example. Solve $\begin{cases} 2x + y = 5 \\ x + 3y = 10 \end{cases}${2x+y=5x+3y=10​ using matrices. In matrix form $A = \begin{pmatrix} 2 & 1 \\ 1 & 3 \end{pmatrix}$A=(21​13​), $\mathbf{b} = \begin{pmatrix} 5 \\ 10 \end{pmatrix}$b=(510​). The determinant is $6 - 1 = 5$6−1=5, so $A^{-1} = \tfrac{1}{5}\begin{pmatrix} 3 & -1 \\ -1 & 2 \end{pmatrix}$A−1=51​(3−1​−12​). Then $\mathbf{x} = A^{-1}\mathbf{b} = \tfrac{1}{5}\begin{pmatrix} 3(5) - 1(10) \\ -1(5) + 2(10) \end{pmatrix} = \tfrac{1}{5}\begin{pmatrix} 5 \\ 15 \end{pmatrix} = \begin{pmatrix} 1 \\ 3 \end{pmatrix}$x=A−1b=51​(3(5)−1(10)−1(5)+2(10)​)=51​(515​)=(13​), so $x = 1$x=1, $y = 3$y=3.

A common pitfall is writing the matrix equation as $\mathbf{x} = \mathbf{b}A^{-1}$x=bA−1, the wrong order — you must pre-multiply by $A^{-1}$A−1, giving $\mathbf{x} = A^{-1}\mathbf{b}$x=A−1b. Another is missing the geometric interpretation of a zero determinant. For the three-variable case and the singular-system analysis, see the Solving Systems with Matrices lesson.

Linear Transformations in 2D

Every $2 \times 2$2×2 matrix represents a linear transformation of the plane — a map that sends the position vector $\begin{pmatrix} x \\ y \end{pmatrix}$(xy​) to $A\begin{pmatrix} x \\ y \end{pmatrix}$A(xy​), keeps the origin fixed, and sends straight lines to straight lines. The key trick for finding the matrix of a known transformation is to track where the unit vectors $\begin{pmatrix} 1 \\ 0 \end{pmatrix}$(10​) and $\begin{pmatrix} 0 \\ 1 \end{pmatrix}$(01​) go: their images become the first and second columns of the matrix.

Transformation	Matrix
Rotation anticlockwise by $\theta$θ about the origin	$\begin{pmatrix} \cos\theta & -\sin\theta \\ \sin\theta & \cos\theta \end{pmatrix}$(cosθsinθ​−sinθcosθ​)
Reflection in the $x$x-axis	$\begin{pmatrix} 1 & 0 \\ 0 & -1 \end{pmatrix}$(10​0−1​)
Reflection in the line $y = x$y=x	$\begin{pmatrix} 0 & 1 \\ 1 & 0 \end{pmatrix}$(01​10​)
Enlargement, scale factor $k$k, centre origin	$\begin{pmatrix} k & 0 \\ 0 & k \end{pmatrix}$(k0​0k​)

The determinant has a clean geometric meaning here: $|\det A|$∣detA∣ is the area scale factor of the transformation, and a negative determinant signals that the transformation reverses orientation (as a reflection does). Composing transformations is just matrix multiplication — but remember the order. To do transformation $P$P first and then $Q$Q, you apply the matrix $QP$QP (the second transformation sits on the left), because it acts on the vector last.

A worked example. Find the single matrix for "reflect in the $x$x-axis, then rotate $90^\circ$90∘ anticlockwise." The reflection is $P = \begin{pmatrix} 1 & 0 \\ 0 & -1 \end{pmatrix}$P=(10​0−1​); the rotation is $Q = \begin{pmatrix} 0 & -1 \\ 1 & 0 \end{pmatrix}$Q=(01​−10​). The combined matrix is $QP = \begin{pmatrix} 0 & -1 \\ 1 & 0 \end{pmatrix}\begin{pmatrix} 1 & 0 \\ 0 & -1 \end{pmatrix} = \begin{pmatrix} 0 & 1 \\ 1 & 0 \end{pmatrix}$QP=(01​−10​)(10​0−1​)=(01​10​), which is reflection in $y = x$y=x.

A common pitfall is composing in the wrong order — for "first $P$P, then $Q$Q" you need $QP$QP, not $PQ$PQ. Another is muddling the signs in the rotation matrix. For the unit-vector method and orientation, see the Linear Transformations in 2D lesson.

Linear Transformations in 3D

The same ideas extend to three dimensions, where $3 \times 3$3×3 matrices transform space. The unit-vector method still works: the images of $\mathbf{i}, \mathbf{j}, \mathbf{k}$i,j,k form the three columns of the matrix. AQA focuses on rotations about the coordinate axes and reflections in the coordinate planes.

For example, a rotation about the $x$x-axis leaves the $x$x-coordinate untouched and rotates the $y$y- and $z$z-coordinates exactly like a 2D rotation:

$R_x(\theta) = \begin{pmatrix} 1 & 0 & 0 \\ 0 & \cos\theta & -\sin\theta \\ 0 & \sin\theta & \cos\theta \end{pmatrix}.$Rx​(θ)=​100​0cosθsinθ​0−sinθcosθ​​.

Reflections in the coordinate planes are even simpler: reflection in the $xy$xy-plane negates the $z$z-coordinate, giving a matrix that is the identity with a $-1$−1 in the bottom-right corner. As in 2D, $|\det|$∣det∣ is the volume scale factor, and a negative determinant means orientation is reversed. Composition is again matrix multiplication in reverse order.

A worked example. Find the image of the point $(1, 2, 3)$(1,2,3) under reflection in the $xy$xy-plane. The matrix is $\begin{pmatrix} 1 & 0 & 0 \\ 0 & 1 & 0 \\ 0 & 0 & -1 \end{pmatrix}$​100​010​00−1​​, so the image is $\begin{pmatrix} 1 \\ 2 \\ -3 \end{pmatrix}$​12−3​​ — the $z$z-coordinate flips sign while $x$x and $y$y are unchanged.

A common pitfall is putting the rotation block in the wrong rows and columns — a rotation about the $x$x-axis fixes the $x$x-row and $x$x-column, with the $\cos/\sin$cos/sin block in the $y,z$y,z positions. Another is dropping a sign in the off-diagonal entries. For the standard 3D matrices and worked images, see the Linear Transformations in 3D lesson.

Eigenvalues and Eigenvectors

Most transformations move and rotate vectors, but for any square matrix there are usually special directions that are merely scaled, not rotated. A non-zero vector $\mathbf{v}$v is an eigenvector of $A$A, with eigenvalue $\lambda$λ, if

$A\mathbf{v} = \lambda \mathbf{v}.$Av=λv.

In words: applying the transformation to an eigenvector just stretches it by the factor $\lambda$λ, leaving its direction unchanged (or reversed, if $\lambda$λ is negative). These directions reveal the deep structure of a transformation, and they are the key to diagonalisation and to computing high powers.

The method is systematic. Rearranging gives $(A - \lambda I)\mathbf{v} = \mathbf{0}$(A−λI)v=0, which has non-zero solutions only when $A - \lambda I$A−λI is singular — that is, when the characteristic equation $\det(A - \lambda I) = 0$det(A−λI)=0 holds. Solve this for the eigenvalues $\lambda$λ. Then, for each eigenvalue, substitute back into $(A - \lambda I)\mathbf{v} = \mathbf{0}$(A−λI)v=0 and solve the resulting system for the corresponding eigenvector (which is determined only up to a scalar multiple).

A worked example. Find the eigenvalues and eigenvectors of $A = \begin{pmatrix} 2 & 1 \\ 1 & 2 \end{pmatrix}$A=(21​12​). The characteristic equation is $\det\!\begin{pmatrix} 2-\lambda & 1 \\ 1 & 2-\lambda \end{pmatrix} = (2-\lambda)^2 - 1 = 0$det(2−λ1​12−λ​)=(2−λ)2−1=0, giving $(2-\lambda)^2 = 1$(2−λ)2=1, so $\lambda = 1$λ=1 or $\lambda = 3$λ=3. For $\lambda = 3$λ=3: $(A - 3I)\mathbf{v} = \begin{pmatrix} -1 & 1 \\ 1 & -1 \end{pmatrix}\mathbf{v} = \mathbf{0}$(A−3I)v=(−11​1−1​)v=0 gives $-x + y = 0$−x+y=0, so an eigenvector is $\begin{pmatrix} 1 \\ 1 \end{pmatrix}$(11​). For $\lambda = 1$λ=1, the same method gives $\begin{pmatrix} 1 \\ -1 \end{pmatrix}$(1−1​).

A common pitfall is solving $\det(A - \lambda I) = 0$det(A−λI)=0 but then forgetting to find the eigenvectors, or quoting $\mathbf{v} = \mathbf{0}$v=0 (which is excluded by definition — eigenvectors are non-zero). Another is sign-slips in expanding the characteristic determinant. For the full method on $2\times 2$2×2 and $3\times 3$3×3 matrices, see the Eigenvalues and Eigenvectors lesson.

Diagonalisation

If a matrix $A$A has enough independent eigenvectors, it can be diagonalised: written as

$A = PDP^{-1},$A=PDP−1,

where $D$D is the diagonal matrix of eigenvalues and $P$P is the matrix whose columns are the corresponding eigenvectors, taken in the same order. Diagonal matrices are wonderfully easy to work with, so diagonalisation turns hard problems into easy ones.

The headline application is computing high powers. Because $A^n = PD^nP^{-1}$An=PDnP−1, and raising a diagonal matrix to a power just raises each diagonal entry to that power, you can find $A^{10}$A10 or $A^{100}$A100 with almost no effort. This is enormously cheaper than multiplying $A$A by itself ten times, and it underlies the analysis of recurrence relations, population models and Markov chains.

A worked example. Using the matrix $A = \begin{pmatrix} 2 & 1 \\ 1 & 2 \end{pmatrix}$A=(21​12​) from above, with eigenvalues $1$1 and $3$3 and eigenvectors $\begin{pmatrix} 1 \\ -1 \end{pmatrix}$(1−1​) and $\begin{pmatrix} 1 \\ 1 \end{pmatrix}$(11​), take $P = \begin{pmatrix} 1 & 1 \\ -1 & 1 \end{pmatrix}$P=(1−1​11​) and $D = \begin{pmatrix} 1 & 0 \\ 0 & 3 \end{pmatrix}$D=(10​03​). Then $A^n = P\begin{pmatrix} 1 & 0 \\ 0 & 3^n \end{pmatrix}P^{-1}$An=P(10​03n​)P−1, and substituting any $n$n gives the power directly.

A common pitfall is mismatching the order of eigenvalues in $D$D with the order of eigenvectors in $P$P — the $k$kth column of $P$P must be the eigenvector for the $k$kth diagonal entry of $D$D. Another is forgetting that diagonalisation needs a full set of independent eigenvectors. For the powers-of-a-matrix method, see the Diagonalisation lesson.

The Cayley-Hamilton Theorem

The Cayley-Hamilton theorem is one of the most elegant results in the matrix course: every square matrix satisfies its own characteristic equation. If the characteristic equation of a $2 \times 2$2×2 matrix is $\lambda^2 - (\operatorname{tr} A)\lambda + \det A = 0$λ2−(trA)λ+detA=0, then substituting the matrix for $\lambda$λ gives the true matrix identity

$A^2 - (\operatorname{tr} A)\,A + (\det A)\,I = O,$A2−(trA)A+(detA)I=O,

where $\operatorname{tr} A$trA is the trace (the sum of the leading-diagonal entries) and $O$O is the zero matrix. The theorem holds for any square matrix, not just $2\times 2$2×2.

This is not merely pretty — it is a labour-saving device. Rearranging the identity expresses $A^2$A2 in terms of $A$A and $I$I, which lets you reduce any power of $A$A to a linear combination of $A$A and $I$I, avoiding repeated multiplication. It also gives a quick route to the inverse: rearranging to make $I$I the subject and factoring out $A$A shows that $A^{-1}$A−1 is a simple linear expression in $A$A, provided $\det A \neq 0$detA=0.

A worked example. For $A = \begin{pmatrix} 1 & 2 \\ 3 & 4 \end{pmatrix}$A=(13​24​), the trace is $1 + 4 = 5$1+4=5 and the determinant is $4 - 6 = -2$4−6=−2, so Cayley-Hamilton gives $A^2 - 5A - 2I = O$A2−5A−2I=O, i.e. $A^2 = 5A + 2I$A2=5A+2I. Rearranging for the inverse, $A(A - 5I) = 2I$A(A−5I)=2I, so $A^{-1} = \tfrac{1}{2}(A - 5I) = \tfrac{1}{2}\begin{pmatrix} -4 & 2 \\ 3 & -1 \end{pmatrix}$A−1=21​(A−5I)=21​(−43​2−1​). This agrees with the standard inverse formula.

A common pitfall is mis-stating the characteristic equation, especially dropping the $\det A$detA constant or getting the sign of the trace term wrong. Another is forgetting that the constant must multiply the identity $I$I, not appear as a bare number, in a matrix equation. For power-reduction and inverse derivations, see the Cayley-Hamilton Theorem lesson.

Common Mark-Loss Patterns Across Matrices

Across the whole topic, a small set of habits accounts for a disproportionate share of lost marks. Most are not about content you do not know — they are about content you do know, applied carelessly under exam pressure.

• Multiplying matrices entrywise instead of row-by-column, or reversing the order of a product.
• Forgetting the alternating sign pattern in a $3 \times 3$3×3 cofactor expansion.
• Forgetting to divide by the determinant when finding an inverse, or dividing only some entries.
• Writing the matrix equation as $\mathbf{x} = \mathbf{b}A^{-1}$x=bA−1 — you must pre-multiply, giving $\mathbf{x} = A^{-1}\mathbf{b}$x=A−1b.
• Composing transformations in the wrong order — for "first $P$P, then $Q$Q," apply $QP$QP.
• Solving the characteristic equation but not finding the eigenvectors, or quoting the zero vector as an eigenvector.
• Mismatching the order of eigenvalues in $D$D with the eigenvectors in $P$P when diagonalising.
• Mis-stating the characteristic equation in Cayley-Hamilton — sign of the trace term, the $\det A$detA constant, and the $I$I factor all matter.
• Treating a singular system as having a unique solution rather than checking for no/infinitely many solutions.
• Not showing enough working. AQA mark schemes award method marks generously when the working is clear; a bare final answer can score fewer marks than a flawed answer with clean method.

A revision plan that explicitly drills these habits — not just the content — will move your grade more than another pass through the textbook.

How LearningBro's AQA Further Maths Matrices Course Helps

LearningBro's AQA A-Level Further Maths: Matrices course is built around the structure of this guide. Each lesson covers one focus area of the compulsory Pure content — operations and multiplication, determinants, inverses, solving systems, transformations in 2D and 3D, eigenvalues and eigenvectors, diagonalisation, and Cayley-Hamilton — with worked examples, mark-scheme-style solutions, and practice questions designed for spaced revisits.

The course is designed to be used in two ways. As a first pass, you can work through the lessons in order, building the structural ideas on top of the algebra and geometry exactly as the subject is best learned. As a revision tool, you can drop into any single lesson — drilling eigenvectors and diagonalisation together before a mock, say — and work the practice independently. The AI tutor is available throughout to give targeted hints when you get stuck, without giving away full solutions, and to mark your written working with structured feedback.

If you want one place to revise matrices well, with realistic practice and clean explanations of every topic, the full course is the right next step. Start with the AQA A-Level Further Maths: Matrices course.

---

Related Reading

• AQA A-Level Further Maths: Complex Numbers Revision Guide
• AQA A-Level Further Maths: Further Calculus Revision Guide
• AQA A-Level Maths: Advanced Algebra Revision Guide