You write a linear equation from a table of values by finding the slope and one point, then assembling the equation in slope-intercept form. Take any two rows, divide the change in y by the change in x, and use that slope with a single ordered pair. When the x-values step evenly, check whether the step from x = 0 appears; that row hands you the y-intercept outright. Here's the full method, the zero-row shortcut, the parallel-error to avoid, and what to do when the table doesn't cooperate.

Read the Table as Two Columns of a Graph

A table of values is a coordinate list wearing a grid. Each row is one ordered pair, (x, y), and each column is one variable. The left column is x, the input; the right is y, the output. When the table is presented vertically with the columns side by side, the leftmost figure in a row is x and the rightmost is y.

Plot the points mentally or on scratch paper before touching algebra. Two points fix any straight line; every additional row is a check that the line stays straight. If the plotted points line up, the data is linear and the equation exists. If they bend, no single linear equation fits, and the honest answer is "not linear" rather than a forced slope.

The slope itself is the change in y divided by the change in x between any two rows. In plain terms, it measures how many units y moves for each single unit x moves. A positive slope rises left to right; a negative slope falls. A slope of zero is a flat line, and a vertical table has no finite slope at all β€” you'll write x equals a constant instead.

The y-intercept is where the line crosses the vertical axis at x = 0. That's why a row with x = 0 in the table is gold: its y-value is the intercept, no calculation needed. Most tables don't include that row, which is why the method below gives you a workaround for when it's missing.

Find the Slope First, Always

Start every problem the same way: extract two rows and compute the slope. The slope is the change in y divided by the change in x between any two points on the line, and it's the same number no matter which pair you pick β€” that's the defining property of a linear relationship.

Take two rows, say (1, 4) and (3, 10). Compute slope as the difference in y over the difference in x: (10 βˆ’ 4) Γ· (3 βˆ’ 1) = 6 Γ· 2 = 3. That 3 is the slope, often written m. Every pair of points in a truly linear table yields that same ratio, so a third pair is your proof.

If the result is negative, keep the sign. A table going from (0, 8) to (4, 0) gives (0 βˆ’ 8) Γ· (4 βˆ’ 0) = βˆ’2, a line falling two units for every one unit right. Sign errors account for most wrong answers on this problem, so write the subtraction in the same order both times: yβ‚‚ βˆ’ y₁ over xβ‚‚ βˆ’ x₁.

Run the slope twice with different row pairs as a self-check. If the two slopes disagree, the table isn't linear and you should stop rather than write a false equation. If they agree, expect that number to hold everywhere in the table, including the row you didn't use.

Use the Zero Row When You Have One

When the table contains a row where x = 0, the y-intercept is that row's y-value β€” read it straight off the grid of numbers. The y-intercept is simply the y-value at x = 0, the point where the line meets the vertical axis, so no algebra is required for that piece.

Suppose a table shows (0, 4), (2, 10), and (4, 16). The zero row says the y-intercept is 4, and the slope from the first two rows is (10 βˆ’ 4) Γ· (2 βˆ’ 0) = 3. Write y = 3x + 4 and the job is done. Two facts, one formula.

This shortcut only works if the zero row truly reads x = 0, not a value that looks close. A row of (0.5, 7) is not the intercept; it's just another point that must go through the slope formula. Double-check the column headers and the units before trusting the shortcut.

Even with the zero row, verify the finish line. Substitute a third point from the table into your equation: (4, 16) into y = 3x + 4 gives 16 = 3(4) + 4. True statements on every row mean the equation is correct, and this one-line check catches sign errors and swapped coordinates.

Substitute When the Table Skips the Zero Row

Most textbook tables skip x = 0, so you compute the intercept with substitution. Use the point-slope form y βˆ’ y₁ = m(x βˆ’ x₁), plug in the slope you found and any row from the table, then solve for y to reach slope-intercept form y = mx + b.

Take a table of (2, 5), (4, 11), and (6, 17). The slope from the first two rows is (11 βˆ’ 5) Γ· (4 βˆ’ 2) = 6 Γ· 2 = 3. Pick (2, 5) and substitute: 5 = 3(2) + b, so 5 = 6 + b, and b = βˆ’1. The equation is y = 3x βˆ’ 1.

Check with a row you didn't use: (6, 17) into y = 3x βˆ’ 1 gives 17 = 18 βˆ’ 1. True. The numbers agree on every row, which is what verifies a correct intercept.

When the arithmetic gets ugly β€” a slope of 2.5 with a point like (4, 3) β€” substitute into slope-intercept form directly rather than point-slope. Solve 3 = 2.5(4) + b to get b = βˆ’7, and the equation reads y = 2.5x βˆ’ 7. Fractions behave the same way: keep the slope as a fraction and let substitution handle the rest.

Work a Full Example From Start to Finish

Here's the whole method on one table so the sequence is unmistakable. Take x of 1, 3, 5, 7 and y of 9, 19, 29, 39. The x-values step by 2 and the y-values step by 10, so the slope is (19 βˆ’ 9) Γ· (3 βˆ’ 1) = 5. The values rise five units for every one step of x.

Pick any row β€” (3, 19) β€” and substitute into slope-intercept form: 19 = 5(3) + b, giving b = 4. The equation is y = 5x + 4. Test it against the row you skipped: x = 5 gives 5(5) + 4 = 29, matching the table. x = 7 gives 39.

The numbers tell a story beyond the grid. At x = 0 the line hits 4, which the table never shows. That intercept is the value of y when x is nothing at all β€” the starting amount before the first step of 1. Framing it that way turns a symbol into a sentence.

If this table described a phone plan, 4 would be the base monthly charge and 5 the cost per unit over the base. That's the actual use of extracting an equation: it lets you predict a value the table never listed, such as the cost at x = 12, which is 64. Prediction is the point.

Handle Fractional and Decimal Slopes

Fractional slopes are the case most learners stall on, and the numbers are fine. Consider x of 2, 6, 10 with y of 1, 4, 7. The slope is (4 βˆ’ 1) Γ· (6 βˆ’ 2) = 3 Γ· 4, so m = 3/4. Substitute (2, 1): 1 = (3/4)(2) + b, so 1 = 1.5 + b and b = βˆ’0.5. The equation is y = (3/4)x βˆ’ 0.5.

Keep the slope as a fraction rather than converting to a decimal unless the decimal terminates cleanly. Three-fourths repeats no digits, but a slope like 1/3 becomes 0.333… and gets rounded in ways that fail a check. Fractions survive checks exactly.

Decimal slopes appear in real data tables and are messy but manageable. A table of (1, 4.2) and (5, 10.2) gives a slope of 6 Γ· 4 = 1.5, and the intercept follows from 4.2 = 1.5(1) + b, so b = 2.7. When the slope has more decimal places, round only at the final step and say so.

Watch for slopes that look fractional because the x-step exceeds 1. A table where x jumps by 3 introduces a deceptive denominator. Take (3, 2) and (9, 8): the slope is 6 Γ· 6 = 1, not 1/3, because both coordinate differences are read from the row pair, not from the x-step alone.

Avoid the Parallel-Table Mistake

The commonest failure on this task is confusing a table with two independent sequences with a linear relationship. Consider a table where the top row is 1, 2, 3, 4 and the bottom row is 2, 4, 6, 8. Both rows increase by 1 and by 2 respectively, and the bottom row is exactly twice the top row. That relationship happens to be linear β€” y = 2x β€” but the reasoning that "the bottom adds 2 each time" only works because the top adds 1 each time.

Change the top row to 1, 3, 5, 7 with the bottom row 2, 6, 10, 14. The bottom adds 4 each time, and the slope is (6 βˆ’ 2) Γ· (3 βˆ’ 1) = 2, not 4. The common difference in the y-column is not the slope unless the x-column steps by exactly 1. That's the trap: row-wise pattern-spotting feels like slope but isn't.

The check that exposes the error is computing slope across two non-adjacent rows. With x of 1, 3, 5, the jump from x = 1 to x = 5 is 10 units in y over 4 units in x, giving 2.5 β€” wait, that disagrees with 2, which means the table isn't the clean line it appeared to be, and you've found an arithmetic slip before it cost you the answer.

A parallel table in which both sequences step by a constant but the ratio of the steps varies is not linear. The slope across each adjacent pair is the change in y over the change in x for that pair, so uneven x-steps demand a per-pair check. Two pairs with the same ratio certify the line; two pairs with different ratios falsify it.

Spot the Special Cases: Horizontal and Vertical Lines

Two cases produce a table that doesn't fit slope-intercept form at all, and both are legitimate answers. A horizontal table has the same y in every row: x of 1, 2, 3 with y of 5, 5, 5. The slope is (5 βˆ’ 5) Γ· (1 βˆ’ 2) = 0, and the equation is y = 5, because y never changes.

A vertical table has the same x in every row: x of 4, 4, 4 with y of 1, 3, 7. The change in x is zero, and division by zero is undefined, so there is no slope. The equation is x = 4. That's a line standing upright, and the slope-intercept form can't represent it.

The distinction matters because the formulas will fail silently otherwise. Many learners compute 0 Γ· 0 or divide by the zero x-difference and write a wrong m. Checking the x-column for repeats and the y-column for repeats takes ten seconds and prevents a wrong answer.

Mixed tables are the failure case worth naming. A row of y = 0 does not make the line horizontal; zero is an ordinary y-value. A row of x = 0 does not make the line vertical; that's the intercept row. Only an entire column frozen at one value triggers these cases.

Verify With a Third Point and a Prediction

Verification is the step that turns an answer into an answer you can defend. Substitute a row you didn't use into the equation you built. Every table row should produce a true statement, and one failure means the slope or intercept is wrong. Do it on every problem where a spare row exists.

A stronger test is prediction. Pick an x-value not in the table and compute the y-value from your equation. Then extend the pattern of the table to see whether the two agree. The equation and the pattern should predict the same number, and when they don't, the table wasn't linear or the equation carries an error.

Take x of 0, 2, 4, 6 and y of 7, 13, 19, 25. Slope is 3, intercept is 7, and the equation is y = 3x + 7. Predict x = 8: the equation gives 31, and extending each column by its step gives 8 and 31. Agreement.

The prediction has a use beyond checking. If the table tracks something real β€” a plant's height over weeks, a tank's depth over minutes β€” the equation tells you when y reaches a value the table never recorded. Set y equal to the target and solve for x. The value of the equation is precisely this reach beyond the recorded rows.

Spot When the Table Isn't Linear

Not every table deserves an equation, and saying so is a correct answer. Compute the slope from the first pair and from the last pair; if the two differ, the relationship is nonlinear. A table of x of 1, 2, 3 and y of 1, 4, 9 is quadratic, not linear, and no straight slope fits it.

A subtler failure is the near-linear table with one bad row. A transcription error or a misread unit can make a single row land off the line, and the honest way to handle it is to report the conflict rather than split the difference. Name which row deviates and present the linear equation the rest of the table supports.

Data taken from a real-world table often rounds, and rounded values can make adjacent slopes disagree in the second decimal. Check whether the differences agree to the precision the table gives. If the table reports whole numbers and the slopes differ by fewer than one part in fifty, rounding explains the gap and the line still fits.

Tables from spreadsheets and lab recorders can carry a repeated x-value from two trials at the same condition. A duplicate x with a different y is not a linear function; it's a set of measurements. Treat it as data with noise, compute the slope by least squares if the task allows, and say plainly that a single straight line through every point does not exist.

FAQ

How do I write the equation when the table's y-intercept isn't shown?

Solve for b with substitution. Compute the slope from any two rows, plug a row's x and y into y = mx + b, and isolate b. A table of (2, 6) and (4, 12) gives a slope of 3, and substituting (2, 6) gives 6 = 6 + b, so b = 0 and the equation is y = 3x.

Does a table with a y-value of zero have an intercept of zero?

No. A y-value of zero at some nonzero x is an x-intercept, not a y-intercept. The y-intercept is the y-value at x = 0 specifically. A table showing (3, 0) tells you the line crosses the horizontal axis at 3, which constrains the equation but doesn't fix b.

What is the slope if the table's x-column never changes?

No finite slope exists. When every row has the same x-value, the run is zero and slope is undefined, so the equation is x equals that constant. A table of x = 5 with y of 2, 6, and 10 represents the vertical line x = 5. Graphing it in a calculator requires menu options outside the standard slope-intercept entry.

Can I use any row from the table for the substitution step?

Yes, any row on the line gives the same b. The arithmetic is easier with a row whose coordinates are small and positive, which lowers the chance of a sign slip. Substitution errors typically appear when the chosen point has a negative coordinate and the sign isn't carried through both terms.

How do I check the equation using only the table?

Substitute every row. Each row produces a statement that must be true if the equation is correct β€” for example, checking (5, 23) against y = 4x + 3 gives 23 = 23. A single false row means a wrong slope or intercept. Two false rows mean the table isn't linear.

Is the answer wrong if I wrote point-slope form instead of slope-intercept form?

Point-slope form y βˆ’ y₁ = m(x βˆ’ x₁) is a correct linear equation for the same line. Grading instructions often require slope-intercept form y = mx + b, so convert at the end. A point-slope answer is algebraically equivalent and passes any check; only the presentation differs.

Why do the y-differences sometimes look like the slope when they aren't?

The y-difference equals the slope only when x-step is 1. A table rising 4 units in y for every 3 units in x has a slope of 4/3, not 4. Always divide by the x-difference, even when that difference looks like an obvious step. The ratio, not the y-change, is the answer.