Exactly one means at least one and at most one
The rule must produce an output for each input it claims to accept, and it must never produce two. Both halves can fail, and “Domain and Range” is largely about the first half.
A function is a rule that assigns to each input exactly one output. The definition is short, the word doing the work is "exactly", and on a graph the whole condition can be checked by eye with the vertical line test.
Core ideas
The first seven lessons used points and lines as solution sets. This lesson names the input-output idea that organises every graph that follows.
The rule must produce an output for each input it claims to accept, and it must never produce two. Both halves can fail, and “Domain and Range” is largely about the first half.
f(x) = x² sends both 3 and −3 to 9, and that is a function. y² = x sends 4 to both 2 and −2, and that is not. The asymmetry is the definition, not a convention.
A vertical line is the set of points sharing one x. If it meets a graph twice, that single input has two outputs. The test is not a separate fact to remember — it is what the definition looks like on paper.
The material
What must hold, what is permitted, and how to check each representation.
A function from a set of inputs to a set of outputs is a rule assigning to each input exactly one output. The input set is the domain; the outputs form the range.
Every word is load-bearing. "Each" forbids an input the rule cannot handle. "Exactly one" forbids an input with two answers. A rule that fails either is not a function, and calling it one anyway breaks everything downstream — notation in “Function Notation” assumes f(3) names a single number, and an inverse in “Inverse Functions” assumes the assignment can be run backwards.
The asymmetry surprises people, so state it plainly: outputs may repeat, inputs may not. Several inputs sharing one output is ordinary and useful. One input with several outputs is a contradiction, because then "the" output of that input does not exist.
The circle is the clearest failure because it fails at almost every input, not at some awkward exception. It is a perfectly good curve and a perfectly good equation; it is simply not a function of x, which is why the course notes list conic sections as out of scope.
A graph in the plane is the graph of a function of x if and only if no vertical line meets it more than once. A vertical line consists of all points with one fixed x, so two intersections are exactly one input with two outputs.
This is where “Graphing a Linear Equation”'s vertical line returns. The graph of x = 5 is itself a vertical line, so a vertical line meets it infinitely often: the input 5 has every number as an output. Of all the lines in this course, the vertical ones are precisely the non-functions — which is the same exception that blocked slope in “Slope” and slope-intercept form in “Slope-Intercept Form”, now explained rather than noted.
y = 2x + 1 function. y = x² function. y = |x| function, since |−7| and |7| are both 7 and each input still gets one output. y² = x not a function. x = 5 not a function. x² + y² = 25 not a function.
One caution about the phrase "function of x". The circle fails as a function of x, and the vertical line x = 5 fails as a function of x while being a perfectly good function of y. The test is always relative to which variable is the input, and the vertical line test presumes the usual choice of x across and y up.
Interactive
The vertical line test, drawn where it fails
Method
Four questions, in this order. The first "no" settles it.
Usually x, but not always. Every later question depends on this choice, so make it explicit before testing anything.
In a table, look for a repeated input. In an equation, solve for the output and watch for a ± or an even root. On a graph, sweep a vertical line.
A rule that says nothing about some input in its stated domain is not a function on that domain. “Domain and Range” turns this into finding the domain.
"Not a function" is half an answer. "Not a function: x = 4 gives both 2 and −2" is the whole one, and it is the form that transfers to every later lesson.
Give the verdict and the witness — the specific input that breaks the rule, when there is one.
Lines 2 and 3 are the ones to dwell on. Both send two different inputs to one output, and both are functions. The rule constrains what leaves an input, not what arrives at an output.
The two tables make the same point in the representation where it is easiest to misread. A repeated row of outputs looks suspicious and is fine; a repeated input with a different output is what disqualifies.
For a faded classification, use the pairs (−2, 1), (0, 3), (2, 1) and (0, 4). The supplied first move is to group them by their first coordinate. Finish the input-output table, give the verdict, and name the exact input and two outputs that witness it. Then remove one pair so the remaining relation becomes a function.
Given equations, tables, sets of pairs and graphs mixed together, you should classify each and, for every failure, produce the input with two outputs. This is the definition the remaining seven lessons build on: notation, domain, transformation, composition and inverse all presuppose that an input has one answer.
Something to change about this lesson? Write it down, tick it off when it is handled, and export the list when you want to act on it. Kept in this browser only: nothing is sent anywhere, and nobody is notified.
No recommendations yet for this lesson.