What is Rank of a Matrix?
To find the rank of a matrix, convert it to row echelon form using row operations and count the number of non-zero rows. This number is the rank.
Think of it as measuring how much “unique information” your matrix contains. If you have a 5×5 matrix but only 3 rows are truly independent, your matrix rank is 3 – the other 2 rows are just combinations of the first three.
Matrix Rank Calculator
Find the Rank using Gaussian Elimination
How do you find the rank of a matrix?
The rank of a matrix is the maximum number of linearly independent rows or columns it contains. To find the rank, convert the matrix to row echelon form (REF) or reduced row echelon form (RREF) using elementary row operations like row swapping, multiplying rows by constants, and adding/subtracting rows. Once in this form, count the number of non-zero rows—this count is the rank. The rank reveals important properties: it indicates the dimension of the column space, tells you how many pivot positions exist, and helps determine if a system of equations has solutions.
How to find the rank of a 3×3 matrix?
To find the rank of a 3×3 matrix, perform row reduction to convert it to row echelon form, then count the non-zero rows.
Steps:
- Apply row operations to create zeros below the diagonal
- Continue until the matrix is in triangular or row echelon form
- Count rows that aren’t all zeros
Example: For a 3×3 matrix, the rank can be 0 (zero matrix), 1 (one independent row), 2 (two independent rows), or 3 (full rank—all rows independent). A full-rank 3×3 matrix is invertible and has a non-zero determinant.
What is the rank of a 2×3 matrix?
A 2×3 matrix (2 rows, 3 columns) can have a rank of 0, 1, or 2 maximum. The rank cannot exceed the smaller dimension, so despite having 3 columns, the maximum rank is limited to 2 by the number of rows.
Possible ranks:
- Rank 0: The zero matrix (all entries are zero)
- Rank 1: Only one linearly independent row
- Rank 2: Both rows are linearly independent (full row rank)
To determine the specific rank of a given 2×3 matrix, perform row reduction and count the non-zero rows in the resulting echelon form.
Formal Definition: The rank of a matrix A is the dimension of the vector space spanned by its rows (or equivalently, by its columns).
What is the Rank of a Matrix?
In formal terms, the rank is the dimension of the vector space generated by the matrix’s rows.
In simple terms, the rank is the count of how many rows remain after you try to eliminate as many of them as possible using subtraction.
- If a row can be made all zeros by subtracting other rows from it, it was Linearly Dependent (redundant).
- If a row cannot be eliminated, it is Linearly Independent (unique).
The Golden Rule
Rank $(A) = $ Number of non-zero rows in Row Echelon Form (REF)
How to Find Rank: 3 Detailed Examples
The best way to learn is by doing. We will use Gaussian Elimination to transform matrices into Row Echelon Form (REF) and then count the survivors.
Example 1: The “Rank Deficient” Matrix
Difficulty: Medium
Let’s find the rank of a $3 \times 3$ matrix where one row is hidden as a combination of the others.
$$A = \begin{bmatrix} 1 & 2 & 3 \\ 2 & 4 & 6 \\ 1 & 1 & 5 \end{bmatrix}$$
Step 1: Eliminate the first column
We want zeros below the top-left $1$.
- Target: The $2$ in Row 2.
- Operation: $R_2 \leftarrow R_2 – 2R_1$
- Calculation: $(2, 4, 6) – 2(1, 2, 3) = (2-2, 4-4, 6-6) = (0, 0, 0)$
- Target: The $1$ in Row 3.
- Operation: $R_3 \leftarrow R_3 – 1R_1$
- Calculation: $(1, 1, 5) – (1, 2, 3) = (0, -1, 2)$
Current State:
$$\begin{bmatrix} 1 & 2 & 3 \\ 0 & 0 & 0 \\ 0 & -1 & 2 \end{bmatrix}$$
Step 2: Organize (Row Swapping)
In Row Echelon Form, rows full of zeros must be at the bottom. Let’s swap Row 2 and Row 3.
- Operation: $R_2 \leftrightarrow R_3$
Final REF:
$$\begin{bmatrix} 1 & 2 & 3 \\ 0 & -1 & 2 \\ 0 & 0 & 0 \end{bmatrix}$$
Step 3: Count Non-Zero Rows
- $[1, 2, 3]$ (Non-zero)
- $[0, -1, 2]$ (Non-zero)
- $[0, 0, 0]$ (Zero)
Result: Since there are 2 non-zero rows, Rank $(A) = 2$.
Example 2: The Rectangular Matrix ($3 \times 4$)
Difficulty: Hard
Rank isn’t just for square matrices. Let’s try a $3 \times 4$ matrix. Remember, the rank cannot exceed the smaller dimension (3).
$$B = \begin{bmatrix} 1 & 1 & 0 & 2 \\ 2 & 1 & 1 & 5 \\ -1 & 0 & 1 & -1 \end{bmatrix}$$
Step 1: Eliminate Column 1
- Target: The $2$ in Row 2.
- Operation: $R_2 \leftarrow R_2 – 2R_1$
- Calculation: $(2,1,1,5) – (2,2,0,4) = (0, -1, 1, 1)$
- Target: The $-1$ in Row 3.
- Operation: $R_3 \leftarrow R_3 + 1R_1$ (We add because the target is negative)
- Calculation: $(-1,0,1,-1) + (1,1,0,2) = (0, 1, 1, 1)$
Current State:
$$\begin{bmatrix} 1 & 1 & 0 & 2 \\ 0 & -1 & 1 & 1 \\ 0 & 1 & 1 & 1 \end{bmatrix}$$
Step 2: Eliminate Column 2
Now we look at the second column, below the diagonal.
- Target: The $1$ in Row 3.
- Operation: $R_3 \leftarrow R_3 + R_2$
- Calculation: $(0,1,1,1) + (0,-1,1,1) = (0, 0, 2, 2)$
Final REF:
$$\begin{bmatrix} 1 & 1 & 0 & 2 \\ 0 & -1 & 1 & 1 \\ 0 & 0 & 2 & 2 \end{bmatrix}$$
Step 3: Count Non-Zero Rows
Every single row still has numbers in it. Nothing was fully eliminated.
Result: There are 3 non-zero rows, so Rank $(B) = 3$. (This is called “Full Row Rank”).
Example 3: The “Tricky” Zero Pivot
Difficulty: Expert
What happens if you have a zero in the pivot position?
$$C = \begin{bmatrix} 0 & 2 & 4 \\ 1 & 3 & 5 \\ 2 & 4 & 2 \end{bmatrix}$$
Step 1: Swap to Fix Pivot
We cannot have a $0$ in the top-left corner. We must swap Row 1 with a row below it that has a non-zero number.
- Operation: $R_1 \leftrightarrow R_2$
Current State:
$$\begin{bmatrix} 1 & 3 & 5 \\ 0 & 2 & 4 \\ 2 & 4 & 2 \end{bmatrix}$$
Step 2: Eliminate Column 1
- Target: The $2$ in Row 3.
- Operation: $R_3 \leftarrow R_3 – 2R_1$
- Calculation: $(2,4,2) – (2,6,10) = (0, -2, -8)$
Current State:
$$\begin{bmatrix} 1 & 3 & 5 \\ 0 & 2 & 4 \\ 0 & -2 & -8 \end{bmatrix}$$
Step 3: Eliminate Column 2
- Target: The $-2$ in Row 3.
- Operation: $R_3 \leftarrow R_3 + R_2$
- Calculation: $(0,-2,-8) + (0,2,4) = (0, 0, -4)$
Final REF:
$$\begin{bmatrix} 1 & 3 & 5 \\ 0 & 2 & 4 \\ 0 & 0 & -4 \end{bmatrix}$$
Result: Even though we had zeros and negative numbers, no row disappeared completely. Rank $(C) = 3$.
Important Rank Properties (Cheat Sheet)
If you are taking an exam, memorize these rules to check your work instantly.
- Rank $\le \min(m, n)$: A $2 \times 5$ matrix can have a rank of at most 2.
- Rank $(A) = \text{Rank}(A^T)$: Transposing a matrix does not change its rank.
- Rank $(AB) \le \min(\text{Rank}(A), \text{Rank}(B))$: Multiplying matrices cannot increase the rank; it can only maintain or decrease it.
- Full Rank: If a square $n \times n$ matrix has Rank $= n$, it is invertible (Non-Singular).
- Rank Deficiency: If Rank $< n$, the determinant is 0 and the matrix is Singular.
Frequently Asked Questions (FAQ)
Q: Can a matrix have a Rank of 0?
A: Yes, but only one matrix in the universe has Rank 0: The Zero Matrix (a matrix filled entirely with zeros). If there is even a single non-zero number, the rank is at least 1.
Q: Does Row Echelon Form always give the unique rank?
A: Yes. While the numbers inside the Row Echelon Form might look different depending on how you swapped rows, the number of non-zero rows will always be the same. The Rank is unique.
Q: What is the relationship between Rank and Solutions?
A:
- Rank $(A) < \text{Rank}(Augmented)$: No Solution (Inconsistent).
- Rank $(A) = \text{Rank}(Augmented) = \text{Variables}$: Unique Solution.
- Rank $(A) = \text{Rank}(Augmented) < \text{Variables}$: Infinite Solutions.
Conclusion
Finding the rank of a matrix is a systematic process of elimination. By using Gaussian Elimination, you peel back the layers of the matrix to see which rows are actually contributing information.
Remember: Rank = Non-Zero Rows in REF.
Don’t want to do the math by hand? Scroll up and use our Matrix Rank Calculator to solve it in seconds.