SAT nonlinear functions: quadratic and exponential
Quadratic and exponential functions change at changing rates. Reading what each number in the rule controls is most of the work.
These are the concepts this skill rests on, the same pages as in the lesson.
Functions, Values, and Zeros
Input In, Output Out
A function turns each input into exactly one output. In , is the output for the input , read "f of x."
Forward: given an input, substitute it for every , in parentheses. . The parentheses matter: , but .
Backward: given an output, set the rule equal to it and solve. means , so , and or . A nonlinear rule can send two inputs to the same output, so a backward question can have two answers; a condition such as picks one.
The same steps work on any rule. For , a negative exponent means a reciprocal: . For , . For , work from the inside out: find first, then put that output into .
Zeros and Intercepts
A zero of is an input that makes the output 0. On the graph of , each zero is where the graph meets the -axis, the -intercept . The -intercept is : set the input to 0.
For , , so the -intercept is . For the zeros, use the fact that a product is 0 only when one of its factors is 0: gives , and gives .
Solve each factor; do not copy the number printed in it. gives , gives 5, gives , and a bare factor gives 0.
A zero is an input; is an output. They answer different questions, so write the point before you answer. A factor that appears twice, as in , gives one distinct zero, 2, and the graph touches the axis there instead of crossing. Some functions have no zeros at all: is never 0.
Quadratic Functions
The Vertex, the Axis, and the Opening
The graph of a quadratic function , with , is a parabola. Its turning point is the vertex, and the vertical line through the vertex is the axis of symmetry: the two sides of the parabola are mirror images across it.
- If , the parabola opens up, and the vertex is its lowest point: the vertex's -coordinate is the minimum value of .
- If , it opens down, and the vertex is its highest point: the -coordinate is the maximum value.
The vertex answers two different questions. Its -coordinate says where the minimum or maximum happens, the input. Its -coordinate says what the minimum or maximum is, the output. Write the vertex as a point, then answer the coordinate the question asks for.
Each Form Shows One Feature
The same quadratic can be written three ways:
| Form | Looks like | Shows, with no work |
|---|---|---|
| standard | the -intercept, | |
| factored | the zeros, and | |
| vertex | the vertex, |
Each form shows its own feature and only that one. Vertex form does not show the -intercept: substitute 0 to get . Standard form does not show the vertex, and factored form does not show the -intercept. All three show , and so whether the parabola opens up or down.
In vertex form, read with its sign changed: is , so . The numbers and are the vertex only when the rule is written exactly as .
When a question asks which equivalent form "displays" a feature as a constant, it is asking which form shows it in this table: the zeros or -intercepts in factored form, a minimum or maximum in vertex form, the -intercept in standard form.
Finding a Hidden Vertex
Standard form and factored form do not show the vertex, so find its -coordinate first. The axis of symmetry of is
and substituting that into the rule gives the vertex's -coordinate.
For : , and . The vertex is , and since , the minimum value is , at .
Why it works: and . The two points at height 5 are mirror images, so the axis is midway between them, at .
Three sign traps: keep the sign of (here is ); divide by , not ; and substitute to get the value. Completing the square gives the same vertex: .
That is one case of a general rule: the axis of symmetry sits midway between any two points of a parabola that are at the same height.
- Zeros and : the axis is .
- Equal outputs, with : the axis is .
For , the zeros are 1 and 5, so the axis is , and the maximum is . The pair gives the same axis.
So factored form gives the vertex in two steps: the midpoint of the zeros, then substitute. For , the axis is and the vertex's -coordinate is , as vertex form showed.
A Missing Constant
When a rule has an unknown constant, each fact you are given makes one equation. Choose the form whose constants the facts fill in.
- A vertex: write with and filled in; one more point gives . With vertex and the point : , so and .
- Two zeros: write ; one more point gives .
- A point at : it hands you in standard form, since .
- An exponential through two points: divide the outputs to find the factor (Linear, Quadratic, or Exponential?).
Then answer what was asked, which is often a value at a different input. Check the finished rule against every fact you were given.
Exponential Functions
Growth and Decay
An exponential function can be written , with and . For these problems, also assume .
- , because . It is the initial value.
- is the growth factor: each time goes up by 1, the output is multiplied by . If , the function grows; if , it decays.
A linear function adds the same amount for each step; an exponential multiplies by the same factor. With , is always positive: the graph comes closer and closer to the -axis on one side but never meets it, so it has no zeros. The line is its horizontal asymptote.
A percent change each period becomes a factor: growth by is , and decay by is . So 4% growth is a factor of 1.04, and 30% decay is a factor of 0.70.
Read a factor back the same way. 1.04 means each period's amount is 104% of the one before: a 4% increase per period. 0.70 means 70% is kept: a 30% decrease. A factor of 2 is a 100% increase. The rate is not the factor: multiplying by 0.04 would leave 4% of the amount, and 1.4 is a 40% increase. A fixed percent each period is exponential; a fixed amount is linear.
A geometric sequence is the same idea counted from 1: if the first term is and each term is times the one before, term is , because the first term has not been multiplied yet. For 3, 12, 48, …, the 5th term is .
The Period Lives in the Exponent
If the factor applies once every units of time, the model is
The exponent counts how many periods have passed.
A population of 200 that doubles every 3 years is . After 6 years, doublings have passed: . Something that halves every 5 hours carries .
The period always matters: 1.04 per year and 1.04 per month are different models. Two common slips are ignoring the period () and multiplying by it ().
Rewriting an Exponential
Exponent rules restate a model without changing a single value. Each rewrite makes a different number visible.
- A new interval. . Times 3 each year is times 9 every 2 years: . In the other direction, a factor of 0.64 every 2 years is each year.
- A new time unit. With in years and in months, , so becomes .
- A hidden coefficient. : , so the initial value is 12, not 3. In , the coefficient is the value at : has the value 7 at .
Linear, Quadratic, or Exponential?
A table tells you its kind of function, but only after you check that the inputs go up in equal steps.
- equal first differences: linear (Skill 03 covers these).
- equal second differences, the differences of the differences: quadratic.
- equal ratios, each output divided by the one before: exponential. The ratio is the factor for one step, and the row at is .
Check at least two differences or two ratios: one is not a pattern. To continue a quadratic table, continue its second differences: for 1, 2, 5, 10, the next first difference is 7, so the next output is 17.
Two points of an exponential, with no row at 0, still give the rule. If , , and , then , so , and . If the inputs go up by 2, the ratio between rows is : take its square root for the factor per 1.
Graphs and Their Changes
Reading a Graph
Read the scale on each axis first: what one gridline is worth. Then read the graph in the same order every time.
- Where it meets the -axis: the zeros. A graph that touches the axis and turns back has a repeated zero there.
- Where it meets the -axis: the -intercept, .
- Where it turns: a maximum or minimum, and the input where it happens.
- Increasing or decreasing: read left to right. Where the graph rises, is increasing; where it falls, decreasing. Name intervals by their -values; an interval that contains a turn is neither.
- The ends: what happens far to the left and right.
To match a graph with its equation, find one decisive feature, such as the -intercept, the direction it opens, a zero, or the turn, and eliminate the choices that fail it. Then check the survivor against a second feature.
Shifts and Reflections
A change to the rule moves the graph. For positive numbers and :
| New rule | Moves the graph of |
|---|---|
| up | |
| down | |
| right | |
| left |
A change outside the function moves the graph the way it reads. A change inside moves it the opposite way: at is , so the point the graph of had at now sits at , 2 units left. Every point moves together, so the vertex of moves to .
A horizontal shift moves every zero by the same amount. A vertical shift does not move the zeros by : it changes which inputs give 0, so solve again.
Keep as your reference, and graph beside it. It starts on top of the reference when and . Set : the second rule becomes , so its vertex moves 2 units left to . Then set : its vertex moves up to .
Two reflections: flips the graph across the -axis, so every output changes sign and a minimum becomes a maximum. flips it across the -axis, so left and right trade places. For an exponential, a shift is also a coefficient: .
Domain, Range, and End Behavior
The domain is the set of inputs a function accepts; the range is the set of outputs it produces.
- Domain. Polynomials and exponentials accept every real number. A square root needs what is inside it to be 0 or more: needs . A denominator cannot be 0: accepts every except 4.
- Range. A parabola's outputs start at its vertex value and run one way: has range if and if ; for example, has range . An exponential with has range : it comes ever closer to but never reaches it.
- End behavior of a polynomial follows its leading term. Even degree: both ends go the same way, up if the leading coefficient is positive. Odd degree: the ends go opposite ways; with a positive leading coefficient the graph falls on the left and rises on the right. So rises on the left and falls on the right.
Models in Context
What Each Number Means
Before reading any number in a model, name the input and the output, each with its unit.
- The coefficient in , and the constant in a standard-form quadratic , are the output at : the starting amount.
- The factor is what the output is multiplied by each period; the percent change per period is .
- A statement says: when the input is , the output is .
- In vertex form , is when (or at what input) the maximum or minimum happens, and is that maximum or minimum. is not the value at .
- A positive zero is the input at which the quantity reaches 0: an object lands, a supply runs out, a profit is $0.
A factor has no unit: in , 1.5 means "50% more each hour," not "1.5 thousand bacteria."
Build a Product Model
When a quantity is the product of two changing lengths, write each length in terms of one variable, then multiply.
A rectangle has a perimeter of 40 meters. If its width is , the width and length together are half the perimeter, 20, so the length is , and the area is . This is factored form: the zeros are 0 and 20, the axis is midway at , and the greatest area is square meters, a 10-meter square.
The situation limits the domain: a width must be positive, and so must the length, so . When you solve an equation from a model, keep only the solutions the situation allows; a negative time or length is rejected (Skill 07 covers the solving). Area and volume formulas are in Skill 17.
Keep going
Build the basics first
Foundations has free readiness checks and lessons in the English and math the SAT expects.
Check Your Skills, the strategy for each kind of question, worked examples and practice from Easy to Hard are in the Quaero SAT Course.
More Advanced Math
Equivalent expressions
See what makes two expressions equivalent, combine like terms without sign errors, expand products, factor, and simplify powers and rational expressions.
Read the concepts →SAT essentials · MathNonlinear equations in one variable
Solve quadratics by factoring, square roots or the quadratic formula, count solutions, and check answers to radical and rational equations.
Read the concepts →SAT essentials · MathSolving formulas for a variable
Rearrange a formula to isolate one letter: undo its operations in order, and handle grouped terms, fractions, squares and square roots.
Read the concepts →SAT essentials · MathSystems of equations in two variables
Solve systems that pair a line with a parabola or a circle, count the points where two graphs meet, and read solutions from a graph or a word problem.
Read the concepts →