Secondary Mathematics: Matrices, Data, Scalar Operations and Matrix Multiplication

Secondary Mathematics · Worked Repair Guide 33

A matrix is a rectangular array of entries with a defined row-and-column structure. It can store data, encode a transformation or represent a relationship between quantities. The arithmetic rules are controlled by that structure: addition compares matching positions, scalar multiplication acts on every entry, and matrix multiplication combines rows with columns in an order-sensitive way.

This guide develops one central habit: check dimensions before operating. Many matrix errors can be rejected before a single product is calculated. Two matrices can be added only when their dimensions match. A product AB can be formed only when the number of columns of A equals the number of rows of B. Even when both AB and BA exist, they need not be equal.

All tables, transitions and numerical examples below are original teaching constructions. This page is a school-level worked-repair route; it does not replace higher linear-algebra or representation-theory owners elsewhere in the BTT Mathematics estate.

1. Read a matrix by rows and columns

Consider A=[[2,5,−1],[4,0,3]]. It has 2 rows and 3 columns, so its dimension is 2×3.

The entry in row 1, column 2 is 5. The entry in row 2, column 3 is 3.

Entry check: a 3×2 matrix has six entries arranged in three rows of two, not two rows of three.

The order “rows × columns” should become automatic because it controls every later compatibility check.

2. Matrix equality is entry-by-entry equality with matching dimensions

Two matrices are equal only when they have the same dimensions and every corresponding entry is equal.

If [[x,2],[5,y]]=[[3,2],[5,−1]], then x=3 and y=−1.

A 1×4 row matrix cannot equal a 4×1 column matrix even if they contain the same four numbers in the same reading order.

Shape is part of the mathematical object.

3. Addition requires identical dimensions

Let A=[[2,−1],[4,3]] and B=[[5,6],[−2,1]]. Then A+B=[[7,5],[2,4]].

Each entry is added to the entry in the same position.

A 2×3 matrix cannot be added to a 3×2 matrix because there is no complete one-to-one position match.

Matrix addition therefore resembles adding tables with the same layout, not pouring all entries into one total.

4. Subtraction also acts position by position

With the same A and B, A−B=[[−3,−7],[6,2]].

Do not subtract only the first row or treat the minus sign as belonging to the matrix name without distributing it to the entries.

Writing A−B=A+(−1)B connects subtraction to scalar multiplication.

This gives a useful check: −B=[[-5,−6],[2,−1]], and adding it to A reproduces the result.

5. Scalar multiplication reaches every entry

If A=[[2,−1],[4,3]], then 3A=[[6,−3],[12,9]].

The scalar 3 multiplies every entry, just as a factor outside an algebraic bracket multiplies every term.

Likewise −2A=[[-4,2],[−8,−6]].

The matrix dimensions do not change under scalar multiplication.

6. Matrix data need labelled meaning

Suppose rows represent two shops and columns represent three products. Matrix Q=[[12,8,5],[9,11,7]] records quantities.

The entry 11 then means “11 units of product 2 at shop 2” under that stated convention.

If the row and column labels are reversed, the same printed numbers describe a different data structure.

A matrix is compact only when its indexing convention remains attached to it.

7. Matrix multiplication is not entry-by-entry multiplication

Let A=[[1,2],[3,4]] and B=[[5,6],[7,8]].

The (1,1) entry of AB comes from row 1 of A and column 1 of B: 1×5+2×7=19.

The complete product is AB=[[19,22],[43,50]].

Multiplying corresponding entries would give [[5,12],[21,32]], which is a different operation and is not ordinary school matrix multiplication.

8. The inner dimensions decide whether a product exists

If A is m×n and B is n×p, then AB exists and has dimension m×p.

The two inner dimensions n must match because each row of A has n entries and each column of B must also have n entries for their dot product.

Example: a 2×3 matrix times a 3×4 matrix gives a 2×4 matrix.

A 2×3 matrix times a 2×4 matrix is not defined in ordinary matrix multiplication because 3≠2.

9. Predict the output size before calculating

If A is 4×2 and B is 2×5, then AB must be 4×5.

This prediction is an inexpensive error check. If a calculation produces a 2×4 or 5×4 result, the multiplication layout has been misread.

The output keeps the number of rows from the first matrix and the number of columns from the second.

A useful memory phrase is “outside dimensions survive”.

10. Work a rectangular product carefully

Let A=[[1,2,3],[0,−1,4]] and B=[[2,1],[3,0],[−1,5]]. A is2×3 and B is3×2, so AB is2×2.

Entry (1,1): 1×2+2×3+3×(−1)=2+6−3=5.

Entry (1,2): 1×1+2×0+3×5=16.

Entry (2,1): 0×2+(−1)×3+4×(−1)=−7.

Entry (2,2): 0×1+(−1)×0+4×5=20.

Therefore AB=[[5,16],[−7,20]].

11. Matrix multiplication is generally not commutative

For A=[[1,2],[3,4]] and B=[[5,6],[7,8]], we found AB=[[19,22],[43,50]].

Now BA=[[23,34],[31,46]].

Thus AB≠BA.

The order matters because row-column pairings change. Never swap matrices merely because ordinary numerical multiplication allows factors to be reversed.

12. Sometimes one order exists and the other does not

If A is2×3 and B is3×4, AB exists and is2×4.

But BA would require the inner dimensions4 and2 to match; they do not. So BA is undefined.

This is stronger than merely saying AB and BA are different.

Matrix order is constrained by dimensions before numerical entries are considered.

13. The identity matrix leaves a compatible matrix unchanged

The 2×2 identity matrix is I=[[1,0],[0,1]].

For any compatible 2×2 matrix A, IA=A and AI=A.

For A=[[2,−1],[4,3]], multiplying shows the rows or columns are reproduced without mixing.

The identity matrix plays a role analogous to the number1 in multiplication, but its size must match the multiplication context.

14. The zero matrix is an additive identity

A zero matrix has every entry equal to0.

For a 2×3 matrix A, adding the 2×3 zero matrix leaves A unchanged.

Multiplying a compatible matrix by a zero matrix produces a zero matrix when the product is defined.

Do not confuse the zero matrix with the scalar0 or with a matrix of the wrong dimensions.

15. Matrix multiplication can aggregate data

Suppose quantity row vector q=[4,3,2] records units of three products, and price column vector p=[[5],[8],[6]] records prices per unit.

The product qp is a1×1 matrix: 4×5+3×8+2×6=56.

This encodes a weighted total.

The order pq would instead be3×3 and represent a different construction, not the same total.

16. A matrix can transform a coordinate vector

As a geometric enrichment example, let T=[[2,0],[0,3]] and column vector v=[[4],[−1]].

Tv=[[8],[−3]]. The x-component has doubled and the y-component has tripled.

The matrix therefore represents a scaling transformation about the origin with different scale factors along the two coordinate axes.

This connects to Transformations, Symmetry, Coordinates and Invariants.

17. Repeated transformation means repeated matrix multiplication

If a transformation matrix T is applied twice, the result is T(Tv)=(T²)v.

For T=[[2,0],[0,3]], T²=[[4,0],[0,9]].

Applying the original scaling twice multiplies x by4 and y by9.

The order of different transformations matters for the same reason matrix multiplication is not generally commutative.

18. Matrix equations can sometimes be solved entry by entry

Suppose X+[[2,−1],[4,3]]=[[7,5],[1,8]].

Subtract the known matrix entry by entry:

X=[[5,6],[−3,5]].

This resembles solving an ordinary equation because matrix addition has an additive inverse and matching dimensions.

More advanced equations involving products need additional structure and should not be treated as ordinary scalar division.

19. Do not divide by a matrix as though it were a number

If AX=B, writing X=B/A is not ordinary matrix notation.

Where inverse matrices are in scope, one may multiply by an inverse under specific conditions, but order matters: from AX=B, left-multiplying by A⁻¹ gives X=A⁻¹B.

Not every square matrix has an inverse.

For a basic Secondary route, the important repair is to avoid inventing scalar-style division rules for matrices.

20. Capstone: quantities, prices and totals

An invented two-branch shop records quantities of three products in Q=[[3,5,2],[4,1,6]]. Price column p=[[7],[4],[5]].

Q is2×3 and p is3×1, so Qp is2×1.

Branch1 total=3×7+5×4+2×5=51.

Branch2 total=4×7+1×4+6×5=62.

Thus Qp=[[51],[62]]. The output dimensions match the interpretation: one total for each of two branches.

21. Independent practice

1. State the dimensions of [[1,2,3],[4,5,6]].
2. Find the entry in row2,column1 of that matrix.
3. If [[x,2],[3,y]]=[[5,2],[3,−4]], find x and y.
4. Add [[1,−2],[4,3]] and [[5,6],[−1,2]].
5. Subtract the second matrix in Question4 from the first.
6. Find −3[[2,−1],[0,4]].
7. Can a2×3 matrix be added to a3×2 matrix? Explain.
8. If A is2×4 and B is4×3, state the dimension of AB.
9. Can BA be formed in Question8? Explain.
10. Multiply [[1,2],[3,4]] by [[2,0],[1,5]].
11. Reverse the order in Question10 and compare.
12. Find [[1,2,3]] [[4],[5],[6]].
13. State the2×2 identity matrix.
14. Find I[[7,−1],[2,5]].
15. Quantity row [4,3,2] and price column [[5],[8],[6]] are multiplied. Find the total.
16. Multiply [[2,0],[0,3]] by [[4],[−1]].
17. Find [[2,0],[0,3]]².
18. Solve X+[[2,−1],[4,3]]=[[7,5],[1,8]].
19. Explain why ordinary entry-by-entry multiplication is not the same as matrix multiplication.
20. Explain why AB need not equal BA.

22. Worked answers

1. 2×3.

2. 4.

3. x=5,y=−4.

4. [[6,4],[3,5]].

5. [[−4,−8],[5,1]].

6. [[−6,3],[0,−12]].

7. No. Addition requires identical dimensions.

8. 2×3.

9. No. B is4×3 and A is2×4, so the inner dimensions3 and2 do not match.

10. [[4,10],[10,20]].

11. [[2,4],[16,22]]. The products differ.

12. [[32]]. 1×4+2×5+3×6.

13. [[1,0],[0,1]].

14. [[7,−1],[2,5]].

15. 56.

16. [[8],[−3]].

17. [[4,0],[0,9]].

18. [[5,6],[−3,5]].

19. Matrix multiplication uses row-by-column sums; entrywise products perform a different operation and generally produce different structural meaning.

20. Changing order changes the row-column pairings and may even make one order undefined.

23. Diagnose matrix errors by checking dimensions before entries

Common failures include reversing rows and columns, adding unlike dimensions, multiplying corresponding entries instead of row by column, predicting the wrong output size, or swapping multiplication order as though matrices were ordinary numbers.

A useful correction note says “rows×columns”, “addition needs same shape”, “inner dimensions match”, or “outside dimensions survive”.

Then ask the learner to state whether the operation is legal before calculating. A correct structural decision is part of the solution, not a preliminary formality.

24. Continue through the BTT learning routes

Return to the BTT Mathematics Hub or BTT Mathematical Lab. Use Transformations, Symmetry, Coordinates and Invariants for geometric transformation structure and Vectors, Magnitude, Direction and Geometric Reasoning for coordinate vectors.

Within Batch09, continue to Sine Rule, Cosine Rule and Non-Right Triangles, Power Functions, Exponential Graphs and Tangent Gradients, or Statistical Diagrams, Stem-and-Leaf, Pie Charts and Misleading Displays.

25. Sources and scope

The matrices, data tables, transformation examples and practice questions are original teaching material. Matrix rules are used in their standard algebraic sense.

For current Singapore Secondary curriculum documents and assessment scope, consult the relevant MOE/SEAB syllabus for the learner’s subject level and year. Match inverse matrices, transformation matrices and repeated products to the actual course being studied.