The range of a linear function read from a graph is every y-value the line touches. Take the lowest point the drawn line actually reaches, take the highest, and report the y-interval between them with open brackets — unless the line runs off the top and bottom of the grid as an arrow-ended line, in which case the range is all real numbers, written (−∞, ∞). The one input you must check first is slope: a nonzero slope guarantees every y-level gets hit at most once, which is what makes the endpoint read-off legitimate. Vertical lines are the exception and are not functions.

Why a Nonzero Slope Makes the Endpoints the Whole Story

A line with any nonzero slope is strictly increasing or strictly decreasing. That single property does the heavy lifting: as x sweeps left to right, y never repeats a value and never turns around, so y can only visit each height once. There is no interior peak, no dip, nothing hiding between the ends. Every range question therefore collapses into two questions — what is the smallest y on the drawn segment, and what is the largest?

Mathematically, this comes from the definition of slope as a constant ratio of change. For y = mx + b with m ≠ 0, rearranging gives x = (y − b)/m, so any chosen y has exactly one matching x. A horizontal line is the degenerate case, and it's worth naming precisely: m = 0 gives y = b for every x, so a horizontal function's range is the single value {b}, often written [b, b].

The graph adds a second constraint that pure algebra misses: domains and ranges from pictures are limited to what is drawn or implied. Textbooks and standardized tests like the College Board's SAT Math section routinely present a segment between two solid dots, and the visual boundary is the intended answer. If the segment is drawn but the equation suggests a full line, trust the drawing unless the problem states the function's domain separately.

One caveat keeps this honest. A straight line with slope zero is still a function, but a vertical line fails the vertical line test, so "range of a vertical line" is a trick question. Its set of y-values is everything, but no function is present to have a range in the first place. Check slope before anything else.

The 60-Second Read-Off: Solid Dots, Open Ends, and One Slope Check

Run this sequence on any straight-line graph, in order. It takes under a minute once you've done it twice.

1: Confirm it's a function. Drag an imaginary vertical line across the graph; if it crosses the drawn line more than once anywhere, stop — you're not looking at a linear function. 2: Identify the endpoints. Look for solid dots (included), open circles (excluded), or arrows at the ends of the line. Arrows mean the line continues indefinitely. 3: Find the lowest y-value. Read it off the y-axis, not the x-axis. On a segment, this is the lower dot's height; on a decreasing arrow-ended line, there is no lowest value, so the low end is −∞. 4: Find the highest y-value. Same logic at the top. On an increasing arrow-ended line, the high end is ∞. 5: Check the slope sign as a sanity test. Rising left to right means the bottom comes from the left endpoint and the top from the right endpoint. Falling means the reverse. If your two readings contradict the slope direction, you've swapped the x- and y-coordinates somewhere. 6: Write the interval with brackets matched to dots. Solid dot → square bracket. Open circle → parenthesis. Arrow → parenthesis on that side. 7: Convert to set-builder if your course demands it. Interval [−2, 5] becomes {y | −2 ≤ y ≤ 5} in the notation most U.S. Algebra I courses use.

Mistakes cluster in the same places. Reading the x-coordinate of an endpoint instead of its y-coordinate is the classic error on a horizontal segment. Excluding an endpoint because the dot "looks" open when it's solid produces a technically wrong interval. And writing a segment's range as (−∞, ∞) because the underlying equation is a full line ignores the drawing entirely — the graph, not the equation, defines what's visible.

Failure handling matters here. If the vertical-line test fails, the picture isn't a function and the whole procedure is inapplicable; say so rather than forcing an interval. If the grid has no scale labels, the answer can only be expressed in terms of the labels that are printed, so state the assumption. And if two dots sit at the same height — a horizontal segment — the range is a single value with both brackets square.

Worked Case One: The Rising Segment That Ends at (−1, 2) and (3, 10)

Picture Graph A: a solid dot at (−1, 2), a solid dot at (3, 10), and a straight segment between them. The slope is (10 − 2)/(3 − (−1)) = 8/4 = 2. Rising line, so the left dot is the bottom and the right dot is the top.

Lowest y is 2, hit at x = −1, and it's included because the dot is solid. Highest y is 10, hit at x = 3, also included. The range is [2, 10]. As an inequality, 2 ≤ y ≤ 10. In set-builder notation, {y | 2 ≤ y ≤ 10}. Nothing in between is missing, because the slope of 2 means y climbs steadily from 2 to 10 as x moves from −1 to 3.

Where students slip on this exact graph is coordinate confusion under time pressure: the point (3, 10) invites a hurried answer of [−1, 3], which is the domain. A useful habit is to physically cover the x-axis while reading, so the eyes can only land on y-values. It costs two seconds and removes the most common error on standardized questions.

Adapt this to any rising segment by remembering the pattern: bottom-left, top-right. A segment rising from (−4, −1) to (2, 5) has range [−1, 5]; a segment rising from (0, −3) to (5, 0) has range [−3, 0]. In every case, compare the two y-coordinates, take the smaller as the floor, and check that both dots are solid.

Slope steepness changes nothing about the method. A line with slope 12 over a short segment produces the same narrow y-interval as a line with slope 1/12 over a long one, provided the y-values of the dots agree. Only the endpoints matter.

Worked Case Two: The Falling Segment From (0, 4) Down to (6, −2)

Graph B flips everything. A solid dot at (0, 4), a solid dot at (6, −2), and a segment connecting them; the slope is (−2 − 4)/6 = −1. Falling line, so the left dot is the top and the right dot is the bottom. The range is [−2, 4].

The failure mode here is mechanical: carrying over the "smallest comes first" choreography from the previous case and pairing 4 with the left bracket. Brackets are ordered low to high by convention, always, regardless of where on the graph the numbers originated. Write the smaller number first even if you read it second.

Errors also appear when the y-values straddle zero, as these do. Some students translate [−2, 4] into 4 ≥ y ≥ −2, which is correct but unconventional; reverse it to −2 ≤ y ≤ 4 for consistency with the interval. If the problem requires set-builder form, use the standard inequality so a grader can't misread the direction.

Falling segments from the opposite quadrant behave identically. A segment from (−3, 1) to (2, −6) has range [−6, 1]. A segment from (−5, 3) to (−1, −2) has range [−2, 3]. Repeat the rule: smaller y first, larger y second, brackets per dot type.

What trips people up more subtly is a falling line whose left endpoint sits low and right endpoint sits high on the page — which can't actually happen; that would be rising. Slope sign and endpoint heights must agree. If a graph's appearance contradicts your computed slope, suspect a mislabeled axis or a misread coordinate before you doubt the rule.

Worked Case Three: The Slope-4 Line Through (−1, 5) With Arrows at Both Ends

Graph C shows a line through (−1, 5) with slope 4, drawn with arrowheads at both the upper-right and lower-left ends. There is no lowest y and no highest y — the line keeps going up forever on the right and down forever on the left. Its range is (−∞, ∞), all real numbers.

This is where students cling to the coordinates they can see. The visible portion of the line between two arrowheads is not the range; it's just a window. If you read the topmost visible pixel and the bottommost visible pixel, you'd get a finite interval that changes as the page is redrawn. The correct answer depends on the arrows, which signal continuation.

Any nonvertical line with arrows at both ends gets this answer regardless of slope: a line of slope 1/4 drawn from (−8, −2) to (8, 2) with arrows also has range (−∞, ∞). Slope magnitude, intercept, and direction are all irrelevant. The arrows do the deciding.

An edge scenario that appears on tests: one arrow only. A ray starting at a solid dot and continuing upward passes a lowest value but never a highest, so its range is [5, ∞) if the dot sits at y = 5. A ray with an open circle at (−1, 5) has range (5, ∞). Read the dot type at the origin of the ray and the arrow at the far end. Two symbols, two brackets.

Vertical lines have arrows and are also infinite in both directions, but as already noted they fail the vertical-line test and aren't functions. If a multiple-choice item offers "(−∞, ∞)" for a vertical line, the intended answer is usually "not a function," and picking the infinity option is a trap.

What Typical Algebra Textbooks and Tests Actually Ask For

The answer format depends on the course level, and getting the format right is sometimes worth as many points as the value. In U.S. Algebra I curricula, including common editions from Pearson and Houghton Mifflin Harcourt, ranges are usually accepted as inequalities or set-builder notation. In precalculus and above, interval notation is the default and set-builder is often optional but expected.

Standardized tests have their own idiom. The College Board's SAT Math presents range questions as multiple-choice items; the answer choice format tells you what's wanted. If the choices are intervals, write intervals; if they're inequalities, write inequalities. Reading the answer choices before computing is a time-saving habit many students skip.

One representation the top search results rarely build is the side-by-side mapping between the four common forms. The table below lays out the same answers across formats.

Graph type Interval notation Inequality Set-builder Number line
Segment, dots at y = 2 and y = 10 [2, 10] 2 ≤ y ≤ 10 {y | 2 ≤ y ≤ 10} Solid dots at 2 and 10, shaded between
Segment, open circle at y = −2, solid at y = 4 (−2, 4] −2 < y ≤ 4 {y | −2 < y ≤ 4} Open dot at −2, solid at 4, shaded between
Horizontal piece at y = 3 [3, 3] y = 3 {y | y = 3} Single solid dot at 3
Ray from solid dot at y = 5 upward [5, ∞) y ≥ 5 {y | y ≥ 5} Solid dot at 5, shaded right
Line with arrows at both ends (−∞, ∞) All real numbers {y | y ∈ ℝ} Full shading

The bracket conventions themselves matter. ∞ and −∞ always take parentheses because they're not numbers you can include. A bracket [∞, something is a notation error that costs points. Similarly, a set-builder expression needs the straight bar, not a colon, in most U.S. textbooks, though some courses accept the colon without penalty.

Common Misreads That Produce the Wrong Answer

The single most frequent error is reporting the x-values. It happens because the leftmost point is easy to spot and inviting, but the question asks about range, which lives on the y-axis. A cheap defense: cover the x-axis with your hand while reading, or trace horizontally from each endpoint over to the y-axis and read there. On a segment drawn through (3, 5) and (7, −1), the range is [−1, 5]; reading the x-coordinates gives [3, 7], which is the domain.

Using parentheses everywhere is the second error. Some students adopt the "always use open intervals" rule from a badly remembered lesson and apply it to solid dots. Solid dot → square bracket; open circle → parenthesis; arrow → parenthesis. There is no ambiguity once you look at the dot. If the dot is missing entirely at one end (a line that fades off the grid without an arrow), check the problem statement; the drawing alone doesn't tell you what's intended.

Ignoring graph boundaries is a third. A segment drawn between two solid dots is finite even if the equation would produce an infinite line. The graph's endpoints govern the answer, and the test-writer drew them deliberately. Conversely, treating arrowheads as decorative and reading off the visible top and bottom of the line produces a finite interval that changes as the grid is rescaled — a subtle but consequential error.

Confusing "no lowest value" with "the lowest is zero" is a fourth. A line that arrows downward has no minimum y, not a minimum of 0. Zero is just a point on the y-axis. The temptation to pick zero comes from the visible bottom of the paper, which isn't a mathematical feature of the line.

Two Independent Sanity Checks: Algebra First, Domain Second

Once you've read an interval off the graph, verify it with the equation if one is given. A segment from (2, −1) to (6, 7) has slope 2, so the equation is y = 2x − 5. Setting y to the low value, −1 = 2x − 5 gives x = 2, matching the endpoint. Setting y to 7 gives x = 6, matching the other endpoint. When the algebra returns the endpoints you read visually, the range is confirmed.

The second check is domain-driven. On segments, the x-interval and y-interval must be "compatible" through the slope. A slope of 2 over an x-interval of width 4 produces a y-interval of width 8, and the segment from (2, −1) to (6, 7) has y-width 8 — consistent. If your endpoints imply a y-width that doesn't match slope times x-width, one of the coordinates was misread.

Neither check is essential on a simple dot-to-dot graph, but both are cheap and catch one-coordinate slips. On exams where each question counts equally, two seconds of verification often prevents a wasted point. Use the checks especially when the y-values are negative and easy to misplace relative to the x-axis.

The width arithmetic is also useful in reverse: given a segment of slope 3 starting at (1, 2) and ending at x = 4, the range's top is 2 + 3 × (4 − 1) = 11, so the range is [2, 11]. That shortcut skips the endpoint coordinates when only one dot is labeled.

When the Range Isn't a Single Interval

Linear functions always give a single-interval range, but the same reading machinery handles the piecewise case. A graph that flattens into a horizontal step and then rises shows a range with a hole: the y-values on the horizontal step are reached, but the step's bottom height equals its top height, so the corresponding height is included only if the step itself is drawn. If the step is at a higher height than where the vertical connector jumps, the y-levels between the two pieces are skipped, and the range splits into two intervals.

Real-world modeling adds a third wrinkle. A linear cost function like y = 12x + 40, drawn on a graph restricted to x between 0 and 10 hours (the modeled shift), has a range of [40, 160] on that restricted domain, even though the formula continues. In admissions, the function y = 4x + 1000 applied to 0 ≤ x ≤ 3 children, exactly as a family works through the cost table, is graphed as a scatter of points; the range is those specific dollars, not all real numbers.

These edge cases rarely appear in an introductory lesson, but the logic is consistent. A segment with one dot solid and the other open uses one square and one round bracket, e.g., (−2, 4] or [−2, 4). A horizontal segment at a y-value between two visible rows counts, and its y-value appears in the range like any other height the line touches.

FAQ

What is the range of y = 3x + 1 graphed as a full line with arrows?

Its range is all real numbers, (−∞, ∞). A line with any nonzero slope hits every real y-value exactly once, because for any target y you can solve for x = (y − 1)/3. The intercept and slope don't limit vertical coverage. Only a horizontal line, or a graph with drawn endpoints, restricts the range.

Does the range depend on the steepness of the slope?

No. A slope of 20 and a slope of 1/20 over the same segment give the same range if the y-coordinates of the endpoints match. Slope affects how quickly y changes, not which heights are reached. A steeper line reaches a given height sooner along the x-axis but still reaches it.

Can the range of a segment include its endpoints exactly?

Yes, if the endpoint is marked with a solid dot. A segment from (1, 2) to (4, 9) with solid dots has range [2, 9], and both 2 and 9 are actual y-values of the function. An open circle at either end changes the bracket to a parenthesis, e.g., (2, 9] or [2, 9). Mixing brackets happens routinely on paired-dot graphs.

What if the line goes below the x-axis and above it?

The procedure is unchanged. Take the smallest y on the graph and the largest y. A segment from (−2, −7) to (3, 5) has range [−7, 5], with 0 sitting inside the interval. Crossing the x-axis is common and doesn't create a discontinuity; the range is still a single interval.

How do I write range in set-builder notation if I learned interval form?

Replace brackets with inequalities. [−2, 5] becomes {y | −2 ≤ y ≤ 5}; (−2, 5) becomes {y | −2 < y < 5}; mixed input maps to the matching inequality or strict inequality at each end. For (−∞, ∞), write {y | y ∈ ℝ} or the equivalent phrase "all real numbers," depending on your course.

Is a horizontal line's range just one number?

Yes. A horizontal line y = 4 has range {4} or [4, 4], because y never varies. It is a valid function with slope zero, and its y-value repeats at every x. Interval notation collapses to a single-point interval with matching brackets.

What if the graph has no scale labels on the y-axis?

Read the answer in terms of the labels that are printed, and state the assumption in your work. If the problem says "each gridline is 2 units," multiply grid steps by 2. If no scale is given at all, the coordinates cannot be determined, and the honest answer is to ask which scale applies or present the range relative to visible tick marks.

Could a range answer include infinity as a bracket?

Never use a square bracket on ∞ or −∞. Those symbols describe unboundedness, not numbers, so they always pair with parentheses: [5, ∞) or (−∞, 3]. Writing [5, ∞] is a notation error, and graders who follow textbook conventions will mark it wrong even though the values themselves are correct.