What Is A Unit Matrix?
A unit matrix is a square matrix in which every entry on the main diagonal is 1 and every other entry is 0. In symbols, for $I = [a_{ij}]$, the entry $a_{ij} = 1$ when $i = j$ and $a_{ij} = 0$ when $i \neq j$.
$$I_3 = \begin{bmatrix} 1 & 0 & 0 \ 0 & 1 & 0 \ 0 & 0 & 1 \end{bmatrix}$$
It is denoted $I$, or $I_n$ when the order $n$ needs to be explicit — so $I_2$ is the $2 \times 2$ unit matrix and $I_4$ is the $4 \times 4$.
Variable glossary. $a_{ij}$ is the entry in row $i$, column $j$. The main diagonal runs top-left to bottom-right, the positions where $i = j$. The subscript $n$ in $I_n$ records the order of the matrix.
Is A Unit Matrix The Same As An Identity Matrix?
Yes — and this is the single most important point of the article. The unit matrix and the identity matrix are the same matrix. "Unit matrix" is an older and more British or Indian-textbook term; "identity matrix" is the more common term in modern and US texts. Both name the square grid of 1s on the diagonal and 0s elsewhere.
So everything true of the identity matrix is true of the unit matrix, with no exceptions:
Multiplying any matrix by it changes nothing: $AI = IA = A$.
Its determinant is 1.
It is its own inverse: $I^{-1} = I$.
Because these properties belong to one object under two names, the identity matrix article carries the full treatment — the determinant proof, the inverse argument, the role in solving $AX = B$. This page exists to settle the terminology and to make the "unit" name mean something, rather than repeat that material.
Why Is It Called A "Unit" Matrix?
The name "unit" is doing real work — it is not arbitrary.
In ordinary arithmetic, the unit is the number 1: the value that leaves any number unchanged under multiplication, since $5 \times 1 = 5$. Mathematicians call 1 the multiplicative identity, or simply the unit.
The unit matrix is the matrix world's version of that exact role. Multiplying any square matrix $A$ by it returns $A$ untouched:
$$A \times I = A$$
So the matrix earns the name "unit" for the same reason the number 1 does — it is the element that does nothing under multiplication. That is also why the alternative name is "identity matrix": it preserves the identity of whatever it multiplies. Two names, one job.
This is the cleanest way to remember which matrix the word points at. A unit matrix is not "a matrix full of units (1s)" — that would be the matrix of all 1s, which does not leave other matrices unchanged. It is the matrix that acts like the number 1.
How Does The Unit Matrix Relate To Other Special Matrices?
The unit matrix sits inside two larger families, which is worth seeing once so the names stop colliding.
It is a diagonal matrix — every off-diagonal entry is 0. A diagonal matrix is the general case; the unit matrix is the special case where every diagonal entry is exactly 1.
It is a scalar matrix — a diagonal matrix with the same value repeated down the diagonal. The unit matrix is the scalar matrix whose repeated value is 1. (A scalar matrix is just $kI$, the unit matrix scaled by $k$ through matrix scalar multiplication.)
Reading those two facts together: every unit matrix is diagonal and scalar, but most diagonal and scalar matrices are not the unit matrix. Only when the shared diagonal value lands on 1 does the matrix become the unit.
Examples Of A Unit Matrix
Example 1
Write the $2 \times 2$ and $3 \times 3$ unit matrices.
Place 1s on the diagonal, 0s elsewhere:
$$I_2 = \begin{bmatrix} 1 & 0 \ 0 & 1 \end{bmatrix}, \qquad I_3 = \begin{bmatrix} 1 & 0 & 0 \ 0 & 1 & 0 \ 0 & 0 & 1 \end{bmatrix}$$
Final answer: the two matrices above.
Example 2
Which of these is a unit matrix? $\begin{bmatrix} 1 & 1 \ 1 & 1 \end{bmatrix}$ or $\begin{bmatrix} 1 & 0 \ 0 & 1 \end{bmatrix}$?
The intuitive reading of "unit matrix" is "a matrix made of units," so the first one — all 1s — looks like the answer.
Wrong attempt. Pick $\begin{bmatrix} 1 & 1 \ 1 & 1 \end{bmatrix}$ because every entry is a 1.
Why it is wrong. Test the defining job: does it leave another matrix unchanged? Multiply it by $\begin{bmatrix} 2 \ 3 \end{bmatrix}$ and you get $\begin{bmatrix} 5 \ 5 \end{bmatrix}$, not $\begin{bmatrix} 2 \ 3 \end{bmatrix}$. It changed the matrix, so it is not the unit.
Correct method. The unit matrix has 1s only on the diagonal. The second matrix, $\begin{bmatrix} 1 & 0 \ 0 & 1 \end{bmatrix}$, satisfies $AI = A$.
Final answer: the second matrix, $\begin{bmatrix} 1 & 0 \ 0 & 1 \end{bmatrix}$.
Example 3
Show that $AI = A$ for $A = \begin{bmatrix} 6 & 1 \ 2 & 8 \end{bmatrix}$.
Multiply $A$ by $I_2$, row against column:
Top-left: $(6)(1) + (1)(0) = 6$
Top-right: $(6)(0) + (1)(1) = 1$
Bottom-left: $(2)(1) + (8)(0) = 2$
Bottom-right: $(2)(0) + (8)(1) = 8$
$$AI = \begin{bmatrix} 6 & 1 \ 2 & 8 \end{bmatrix} = A$$
Final answer: $AI = A$, confirming the unit-matrix property.
Example 4
The matrix $\begin{bmatrix} p & 0 \ 0 & q \end{bmatrix}$ is the $2 \times 2$ unit matrix. Find $p$ and $q$.
For the unit matrix, both diagonal entries must equal 1:
$p = 1$
$q = 1$
Final answer: $p = 1$ and $q = 1$.
Example 5
Find the determinant of the $4 \times 4$ unit matrix.
The unit matrix is diagonal, so its determinant is the product of its diagonal entries:
$|I_4| = 1 \times 1 \times 1 \times 1$
$|I_4| = 1$
Final answer: $|I_4| = 1$, and $|I_n| = 1$ for every order $n$.
Example 6
Is the unit matrix its own inverse?
A matrix times its inverse gives the unit matrix itself. Test $I \times I$ for $I_2$:
$$\begin{bmatrix} 1 & 0 \ 0 & 1 \end{bmatrix}\begin{bmatrix} 1 & 0 \ 0 & 1 \end{bmatrix} = \begin{bmatrix} 1 & 0 \ 0 & 1 \end{bmatrix} = I$$
Since $II = I$, the inverse of $I$ is $I$.
Final answer: Yes, $I^{-1} = I$.
Why The Unit Matrix Matters: "The number 1, rebuilt for matrices"
The unit matrix exists because matrix multiplication needed a "do-nothing" element to make its algebra complete — the same role the number 1 plays for ordinary multiplication. Without it, you could not define what an inverse is, because an inverse is defined as the matrix that multiplies back to the unit.
Where that role shows up:
Defining inverses. The invertible matrix condition is $A A^{-1} = I$. The unit matrix is the target that gives "inverse" its meaning.
Solving systems. Gaussian elimination drives a coefficient matrix toward the unit matrix; the moment the left side becomes $I$, the right side holds the solution.
Resetting transformations. In computer graphics, the unit matrix is the transformation that leaves an object exactly where it is — pipelines reset to it before applying a new rotation or scale.
Students who store "unit matrix" and "identity matrix" as two separate facts waste effort and second-guess themselves on exams when only one name appears. The fix is to read the word "unit" as a pointer to the number 1's job. At Bhanzu, trainers teach the matrix through that job — the element that changes nothing — so the two names land as one idea.
Where Students Slip On The Unit Matrix (And How To Fix It)
Mistake 1: Reading "unit matrix" as "a matrix of all 1s"
Where it slips in: Meeting the term for the first time.
Don't do this: Write $\begin{bmatrix} 1 & 1 \ 1 & 1 \end{bmatrix}$ when asked for the unit matrix.
The correct way: The 1s sit only on the diagonal. The first-instinct reading takes "unit" to mean "full of 1s," but the name refers to the number 1's role (leaving things unchanged), not to filling every cell with a 1. A matrix of all 1s changes whatever it multiplies, so it cannot be the unit.
Mistake 2: Treating unit and identity as different matrices
Where it slips in: Seeing both terms in the same course or comparing two textbooks.
Don't do this: Hunt for a difference between the unit matrix and the identity matrix.
The correct way: They are the same object. The second-guesser who has just learned "identity matrix" and then meets "unit matrix" on a different page assumes the new word must name something new — it does not. Same grid, same properties, two regional names.
Mistake 3: Calling any scalar matrix the unit matrix
Where it slips in: Seeing a clean diagonal matrix with a repeated value.
Don't do this: Call $\begin{bmatrix} 5 & 0 \ 0 & 5 \end{bmatrix}$ the unit matrix because it looks tidy and symmetric.
The correct way: The repeated diagonal value must be exactly 1. The matrix above is a scalar matrix ($5I$), not the unit matrix. The memorizer who recalls "unit matrix is diagonal" drops the second half of the rule: the diagonal value, not just the shape, has to be 1.
Key Takeaways
A unit matrix is a square matrix with 1s on the main diagonal and 0s elsewhere, written $I$ or $I_n$.
"Unit matrix" and "identity matrix" are two names for the same object; the full property treatment lives in the identity matrix article.
The name "unit" points to the number 1's role — the element that leaves anything unchanged under multiplication, since $AI = IA = A$.
The unit matrix is a special diagonal and scalar matrix, the case where the diagonal value is exactly 1.
It is not a matrix of all 1s, and not every scalar matrix is the unit matrix.
A Practical Next Step
Practice these to lock in the equivalence, then read the identity matrix article for the full set of properties, the determinant proof, and the inverse argument:
Write $I_2$, $I_3$, and $I_5$ from memory.
Confirm $AI = A$ for a $3 \times 3$ matrix of your choice.
Decide which of $\begin{bmatrix} 1 & 1 \ 0 & 1 \end{bmatrix}$ and $\begin{bmatrix} 1 & 0 \ 0 & 1 \end{bmatrix}$ is the unit matrix, and say why.
To build matrix fundamentals with a teacher who untangles the terminology for you, work with an algebra tutor, join algebra classes, or try math classes online. Want a live Bhanzu trainer to walk through more matrix problems? Book a free demo class.
Read More
Multiplication of matrices — the operation the unit matrix acts as the identity for.
Matrices and determinants — the wider context, including why the unit matrix has determinant 1.
Solve matrices — where reaching the unit matrix is the goal of elimination.
Symmetric matrix — another named family the diagonal-and-scalar discussion connects to.
Was this article helpful?
Your feedback helps us write better content