Mathwords logoMathwords

Row Equivalence — Definition, Formula & Example

Row equivalence is the relationship between two matrices of the same size when you can turn one into the other using only elementary row operations: swapping rows, scaling a row by a nonzero number, or adding a multiple of one row to another.

Two m×nm \times n matrices AA and BB are row equivalent, written A∼BA \sim B, if there is an invertible m×mm \times m matrix PP with B=PAB = PA. Equivalently, PP is a product of elementary matrices. Row equivalence is an equivalence relation. Each equivalence class contains exactly one matrix in reduced row-echelon form (RREF). So A∼BA \sim B if and only if rref⁡(A)=rref⁡(B)\operatorname{rref}(A) = \operatorname{rref}(B).

Key Formula

B=EkEk−1⋯E1A=PAB = E_k E_{k-1} \cdots E_1 A = PA
Where:
  • A,BA, B = Two matrices of the same size, m by n
  • E1,…,EkE_1, \dots, E_k = Elementary matrices, one for each row operation, applied in order from E1E_1 to EkE_k
  • PP = The invertible m by m product of the elementary matrices

Worked Example

Problem: Decide whether A=[123247]A = \begin{bmatrix} 1 & 2 & 3 \\ 2 & 4 & 7 \end{bmatrix} and B=[241121]B = \begin{bmatrix} 2 & 4 & 1 \\ 1 & 2 & 1 \end{bmatrix} are row equivalent.
Step 1: Reduce AA to reduced row-echelon form.
[123247]→R2−2R1[123001]→R1−3R2[120001]\begin{bmatrix} 1 & 2 & 3 \\ 2 & 4 & 7 \end{bmatrix} \xrightarrow{R_2 - 2R_1} \begin{bmatrix} 1 & 2 & 3 \\ 0 & 0 & 1 \end{bmatrix} \xrightarrow{R_1 - 3R_2} \begin{bmatrix} 1 & 2 & 0 \\ 0 & 0 & 1 \end{bmatrix}
Step 2: Reduce BB the same way. Use R1↔R2R_1 \leftrightarrow R_2, then R2−2R1R_2 - 2R_1, then −R2-R_2, then R1−R2R_1 - R_2.
[241121]→[121241]→[12100−1]→[121001]→[120001]\begin{bmatrix} 2 & 4 & 1 \\ 1 & 2 & 1 \end{bmatrix} \to \begin{bmatrix} 1 & 2 & 1 \\ 2 & 4 & 1 \end{bmatrix} \to \begin{bmatrix} 1 & 2 & 1 \\ 0 & 0 & -1 \end{bmatrix} \to \begin{bmatrix} 1 & 2 & 1 \\ 0 & 0 & 1 \end{bmatrix} \to \begin{bmatrix} 1 & 2 & 0 \\ 0 & 0 & 1 \end{bmatrix}
Step 3: The two RREFs match, so A∼BA \sim B. To build an explicit path from AA to BB, first apply AA's two operations. Then undo BB's operations in reverse order, replacing each one with its inverse: R1+R2R_1 + R_2, then −R2-R_2 (its own inverse), then R2+2R1R_2 + 2R_1, then R1↔R2R_1 \leftrightarrow R_2.
rref⁡(A)=rref⁡(B)=[120001]\operatorname{rref}(A) = \operatorname{rref}(B) = \begin{bmatrix} 1 & 2 & 0 \\ 0 & 0 & 1 \end{bmatrix}
Answer: Yes. AA and BB are row equivalent because they share the same reduced row-echelon form.

Why It Matters

Gaussian elimination works because row-equivalent augmented matrices represent systems with exactly the same solution set. In a first linear algebra course, row equivalence also gives the standard invertibility test: a square matrix AA is invertible if and only if A∼InA \sim I_n.

Common Mistakes

Mistake: Assuming row-equivalent matrices share everything, including their column space or determinant.
Correction: Row operations preserve the row space, the null space, and the rank. They can change the column space and the determinant. For example, [10]\begin{bmatrix} 1 \\ 0 \end{bmatrix} and [01]\begin{bmatrix} 0 \\ 1 \end{bmatrix} differ by one row swap, yet their column spaces are different lines.

Related Terms