Visual Tools
Calculators
Tables
Mathematical Keyboard
Converters
Other Tools

How to Find the Reduced Row Echelon Form of a Matrix

Reduced row echelon form (RREF) is echelon form with two extra conditions: every pivot is 11, and it is the only nonzero entry in its column. Every matrix has exactly one RREF, so you can compare answers entry by entry.

Recognize the task

A question that says "find the RREF", "row-reduce completely" or "reduce to reduced echelon form"; a matrix whose pivot and free columns you need; a test of invertibility (the RREF of a square matrix is II exactly when the matrix is invertible). Test: is a pivot different from 11, or is there a nonzero entry above a pivot? Then you are not done.
The two extra conditions are in the lesson section Reduced Row Echelon Form.

Steps

StepOn (12−1424151221)\begin{pmatrix} 1 & 2 & -1 & 4 \\ 2 & 4 & 1 & 5 \\ 1 & 2 & 2 & 1 \end{pmatrix}
1Reduce to echelon form. Follow row echelon form: pivots and zeros below them.R2−2R1, R3−R1, R3−R2R_2 - 2R_1,\ R_3 - R_1,\ R_3 - R_2: (12−14003−30000)\begin{pmatrix} 1 & 2 & -1 & 4 \\ 0 & 0 & 3 & -3 \\ 0 & 0 & 0 & 0 \end{pmatrix}
2Scale every pivot row so its pivot is 1. Divide the row by its pivot.Elementary row operations ↗13R2\frac{1}{3}R_2: [0 0 1 −1][0\ 0\ 1\ {-1}]
3Clear above each pivot, from the bottom pivot up. Subtract multiples of the pivot row from the rows above it.Elementary row operations ↗R1+R2R_1 + R_2: [1 2 0 3][1\ 2\ 0\ 3]
4Check both conditions. Each pivot is 11 and alone in its column. Columns without a pivot may hold any numbers.(1203001−10000)\begin{pmatrix} 1 & 2 & 0 & 3 \\ 0 & 0 & 1 & -1 \\ 0 & 0 & 0 & 0 \end{pmatrix}, pivot columns 1 and 3

Worked examples

Example 1 · an invertible matrix

2 × 2
(2614)\begin{pmatrix} 2 & 6 \\ 1 & 4 \end{pmatrix}, 12R1\frac{1}{2}R_1: [1 3][1\ 3]
pivot to 1
R2−R1R_2 - R_1: [0 1][0\ 1]; R1−3R2R_1 - 3R_2: [1 0][1\ 0]
clear below, then above
RREF⁡=I\operatorname{RREF} = I
so the matrix is invertible

Example 2 · a wide matrix

2 × 3
(1352712)\begin{pmatrix} 1 & 3 & 5 \\ 2 & 7 & 12 \end{pmatrix}, R2−2R1R_2 - 2R_1: [0 1 2][0\ 1\ 2]
R1−3R2R_1 - 3R_2: (10−1012)\begin{pmatrix} 1 & 0 & -1 \\ 0 & 1 & 2 \end{pmatrix}
clear above the second pivot

Example 3 · a singular matrix

rank 2
(123456789)\begin{pmatrix} 1 & 2 & 3 \\ 4 & 5 & 6 \\ 7 & 8 & 9 \end{pmatrix}, R2−4R1, R3−7R1R_2 - 4R_1,\ R_3 - 7R_1
rows [0 −3 −6][0\ {-3}\ {-6}], [0 −6 −12][0\ {-6}\ {-12}]
R3−2R2R_3 - 2R_2: [0 0 0][0\ 0\ 0]; −13R2-\frac{1}{3}R_2: [0 1 2][0\ 1\ 2]
zero row, pivot to 1
R1−2R2R_1 - 2R_2: (10−1012000)\begin{pmatrix} 1 & 0 & -1 \\ 0 & 1 & 2 \\ 0 & 0 & 0 \end{pmatrix}
not II: not invertible
Check a reduced form in the matrix calculator.

See it happen

In reduced form every pivot column is a column of the identity matrix, so each pivot row says how one pivot unknown depends on the free ones and nothing else. Different sequences of row operations can give different echelon forms, but they all end at the same reduced form, because the reduced form records only the relations between the columns, not the path taken.

Common mistakes

Leaving a pivot that is not 1

✗ (1203003−3)\begin{pmatrix} 1 & 2 & 0 & 3 \\ 0 & 0 & 3 & -3 \end{pmatrix} called RREF
✓ divide row 2 by 33: [0 0 1 −1][0\ 0\ 1\ {-1}]

Clearing only below the pivots

✗ (135012)\begin{pmatrix} 1 & 3 & 5 \\ 0 & 1 & 2 \end{pmatrix} called RREF
✓ clear the 33 above the second pivot: (10−1012)\begin{pmatrix} 1 & 0 & -1 \\ 0 & 1 & 2 \end{pmatrix}

Trying to clear a column with no pivot

✗ trying to make the 22 in [1 2 0 3][1\ 2\ 0\ 3] zero
✓ column 2 has no pivot; its entries stay as they are