Algebra › Section D · Systems of linear equations › Topic D.2
Systems by elimination
Graphing estimates and substitution rewrites; elimination attacks. Adding or subtracting whole equations cancels one variable outright, which is often the fastest route when nothing is easy to isolate.
The third method
A linear system can be solved three ways. Graphing shows the answer but only reads it off a picture, and substitution rewrites one equation in terms of one variable. Elimination takes a third road: if both equations are true, then adding them gives another true statement, by the addition property of equality — and if the terms are arranged so one variable cancels, the sum collapses to a single equation with a single unknown. No isolating, no fractions from dividing.
Adding when coefficients are opposites
Take \(x + y = 12\) and \(x - y = 4\). The y-terms are opposites: \(+y\) and \(-y\). Adding the equations eliminates y in one step: \(2x = 16\), so \(x = 8\). Back-substitute into the first equation: \(8 + y = 12\), so \(y = 4\). Checking in both originals: \(8 + 4 = 12\) and \(8 - 4 = 4\). The solution is the point \((8, 4)\) where the two lines meet.
Subtracting when coefficients are equal
When the matching coefficients are equal rather than opposite, subtract instead. From \(5x + 2y = 19\) and \(5x - 4y = 7\), subtracting the second from the first kills the x-terms: \(6y = 12\), so \(y = 2\), and then \(5x + 4 = 19\) gives \(x = 3\). Subtracting an entire equation means negating every one of its terms, not just the first: the \(-4y\) becomes \(+4y\) here. A reliable habit is to write the second equation underneath with all signs flipped, then add.
Multiplying to match
Most systems do not arrive with matching coefficients, so create them. Multiply an entire equation — every term — by a constant chosen to give a common target, usually the least common multiple of the two coefficients. To eliminate y in \(2x + 3y = 8\) and \(x + 6y = 13\), double the second: \(2x + 12y = 26\). Now the x-terms are equal, so subtract: \(-9y = -18\), giving \(y = 2\), and then \(x = 1\). When neither equation alone can match the other, scale both, each by its own multiplier.
Worked example: scaling both equations
Solve \(3x + 2y = 12\) and \(4x + 3y = 17\). Neither y-coefficient divides the other, so match them at the least common multiple, 6: triple the first to get \(9x + 6y = 36\), double the second to get \(8x + 6y = 34\). The y-terms are now equal, so subtract: \(x = 2\). Back-substitute into the first: \(6 + 2y = 12\), so \(y = 3\). Checking both originals: \(3(2) + 2(3) = 12\) and \(4(2) + 3(3) = 17\). The solution is \((2, 3)\), the single point where the lines cross.
Worked example: when everything cancels
Solve \(3x - 6y = 12\) paired with \(x - 2y = 3\). Triple the second: \(3x - 6y = 9\). Subtracting the first leaves \(0 = 3\), a false statement. No pair \((x, y)\) satisfies both equations, so the system has no solution: the lines are parallel. Now pair the first with \(x - 2y = 4\) instead. Tripling gives \(3x - 6y = 12\), identical to the first equation, and subtracting leaves \(0 = 0\), true for every point on the line. The two equations describe the same line, so there are infinitely many solutions. Elimination exposes both cases automatically.
Choosing the most convenient method
All three methods give the same answer, so pick the one needing the least arithmetic. Graphing suits small-integer problems where seeing the crossing matters. Substitution wins when one equation already isolates a variable, or has a variable with coefficient 1, so isolation is cheap. Elimination wins when both equations sit in standard form \(Ax + By = C\) with awkward coefficients that would turn substitution into a fraction-fest. And elimination is the method that extends unchanged to three variables, and to the matrix methods beyond them.
Back-substitute, then check both
Elimination finds one variable; the job is half done. Back-substitute into whichever original equation makes the arithmetic gentlest — usually the one with the smaller coefficients — and solve for the second variable. Then verify the finished pair in both original equations, not just the one used for back-substitution. A pair that checks in only one equation is the pair to a different system. The check costs seconds and catches sign slips that would otherwise ship.
Common mistakes to avoid
First, subtracting an equation term-by-term without flipping every sign — write the flipped equation underneath and add instead. Second, scaling only part of an equation: the multiplier hits every term on both sides. Third, eliminating before aligning: both equations should be in standard form with like terms stacked in columns. Fourth, announcing \(x\) as the solution: the answer is an ordered pair. Fifth, misreading the special cases — \(0 = k\) with \(k \neq 0\) means no solution, while \(0 = 0\) means infinitely many, not a mistake.
Key takeaways
Elimination adds or subtracts whole equations so one variable cancels: add when matching coefficients are opposites, subtract when they are equal, and when they match in neither way, scale one or both equations to a common coefficient first. Special cases announce themselves: a false statement like \(0 = 3\) means parallel lines with no solution, and \(0 = 0\) means one line written twice, infinitely many solutions. Back-substitute for the second variable and check the pair in both original equations. Choose graphing for small numbers and a picture, substitution when a variable is easy to isolate, and elimination when standard-form coefficients are awkward — every correct method lands on the same point.
Flashcards
Flashcard
Add or subtract equations to cancel a variable
Test your understanding
Question 1 of 5
Solve by adding the equations: x + y = 12 and x - y = 4.
Question 2 of 5
Subtracting x + 6y = 13 (after doubling it) from 2x + 3y = 8 — that is, 2x + 3y = 8 minus 2x + 12y = 26 — gives which equation?
Question 3 of 5
Solve by elimination: 2x + 3y = 11 and 3x - 4y = 8.
Question 4 of 5
After eliminating both variables you are left with 0 = 5. What does the system have?
Question 5 of 5
Which method is most convenient for y = 3x - 1 and x = 5?
