Chapter 3: Linear Functions

Chapter 3 Learning Objectives:

 Use function notation to represent the relationship between inputs and outputs
 Read function inputs and outputs from a graph, table, equation, or word problem
 Recognize a linear function, and identify the slope and initial value
 Calculate a slope and interpret it as a rate of change
 Use technology to create a scatterplot and find a linear regression equation
 Interpret the meaning of the slope and initial value of a regression equation
 Use a regression equation to predict the output for a given input
 Use technology to calculate the correlation
 Interpret the correlation from a value of r, or by looking at a scatterplot

Section 3.1 – Functions

Section 3.2 - Linear Functions
Section 3.3 – Linear Regression
Section 3.4 – Correlation

Section 3.1 – Functions

The natural world is full of relationships between quantities that change. When we see
these relationships, it is natural for us to ask “If I know one quantity, can I then determine
the other?” This establishes the idea of an input quantity, or independent variable, and a
corresponding output quantity, or dependent variable. From this we get the notion of a
functional relationship in which the output can be determined from the input.

For some quantities, like height and age, there are certainly relationships between these
quantities. Given a specific person and any age, it is easy enough to determine their
height, but if we tried to reverse that relationship and determine age from a given height,
that would be problematic, since most people maintain the same height for many years.

A function is a rule for a relationship between an input, or independent, quantity and an

output, or dependent, quantity in which each input value uniquely determines one output
value. We say “the output is a function of the input.”

Function Notation
To simplify writing out expressions and equations involving functions, a simplified
notation is often used.
Chapter 3: Linear Functions 2

FUNCTION NOTATION is used to indicate a functional relationship between two

quantities as follows:
Function Name (INPUT) = OUTPUT

So, the statement f (x) = y would refer to the function f , and correspond to the
ordered pair (x,y), where x is the input variable, and y is the output variable.

Rather than write “height is a function of age”, we could use the descriptive variable h to
represent height and we could use the descriptive variable a to represent age.

“Height is a function of age”

If we name the function f we write “h is f of a”
or more simply “h = f(a)”

We could instead name the function h and write “h(a)”

which is read “h of a” and still means “height is a function of age”.

Remember we can use any variable to name the function; the notation h(a) shows us that
h depends on a. The value “a” must be put into the function “h” to get a result. Be
careful - the parentheses indicate that age is input into the function (Note: do not confuse
these parentheses with multiplication!).

Example 1
The number of days in a month is a function of the name of the month, so if we name the
function f, we could write “days = f(month)” or d = f(m). If we simply name the function
d, we could write d(m)

For example, d(March) = 31, since March has 31 days. The notation d(m) reminds us that
the number of days, d is dependent on the name of the month, m. Here the number of
days is the output, and the name of the month is the input.

Example 2
A function P(n) gives the number of police officers in a town n years after 2000. What
does P(5) = 300 tell us?

When we read P(5) = 300, we see the value for the input quantity of the function is 5,
meaning 5 years after 2000. This gives us the year 2000 + 5 = 2005. The output value is
300, the number of police officers.

So this tells us that in the year 2005 there were 300 police officers in the town.
Chapter 3: Linear Functions 3

When we work with functions, there are two typical things we do: evaluate and
solve. Evaluating a function is what we do when we know an input, and use the
function to determine the corresponding output. Evaluating will always produce one
result, since each input of a function corresponds to exactly one output.

Solving equations involving a function is what we do when we know an output, and

use the function to determine the inputs that would produce that output. Solving a
function could produce more than one solution, since different inputs can produce the
same output.

Example 3
Using the table shown, a) Evaluate g(3) b) Determine n when g(n) = 6.

n 1 2 3 4 5
g(n) 8 6 7 6 8

a) Evaluating g(3) (read: “g of 3”) means that we need to determine the output value of
the function g given the input value of n=3. Looking at the table, we see the output
corresponding to n=3 is g(n)=7, allowing us to conclude g(3) = 7.

b) Determine n when g(n) = 6. Here, we need to determine what input (n) value(s)
produce an output value of 6. Looking at the table we see there are two solutions: n
= 2 and n = 4. When we input 2 into the function g, our output is 6. When we input
4 into the function g, our output is also 6.

You Try 3.1A

The function k(m) is given by the table below.

m -2 0 1 4 8 15
k(m) 4 8 32 51 33 1

a. Evaluate k(4). b. Determine m when k(m)=8.

Graphs as Functions

Oftentimes a graph of a relationship can be used to define a function. By convention,

graphs are typically created with the input quantity along the horizontal axis and the
output quantity along the vertical axis. The most common graph has y on the vertical
axis and x on the horizontal axis, and we say y is a function of x, or y = f(x) when the
function is named f.

Evaluating a function using a graph requires taking the given input and using the graph to
look up the corresponding output. Solving a function equation using a graph requires
taking the given output and looking on the graph to determine the corresponding input.
Chapter 3: Linear Functions 4

Example 4
Given the graph below, a) Evaluate f(2) b) Determine x when f(x) = 4

a) To evaluate f(2), we find the input of x=2 on the horizontal axis. Moving up to the
graph gives the point (2, 1), giving an output of y=1. So f(2) = 1
b) To determine x when f(x) = 4, we find the value 4 on the vertical axis because if
f(x) = 4 then 4 is the output. Moving horizontally across the graph gives two points
with the output of 4: (-1,4) and (3,4). These give the two solutions to f(x) = 4:
x = -1 or x = 3. This means f(-1)=4 and f(3)=4, or when the input is -1 or 3, the
output is 4.

You Try 3.1B

The function f(x) is given below.

a. Evaluate f(1)

b. Determine x when f(x)=1.

Formulas as Functions

When possible, it is very convenient to define relationships using formulas. If it is

possible to express the output as a formula involving the input quantity, then we can
define a function. As with tables and graphs, it is common to evaluate and solve
functions involving formulas. Evaluating will require replacing the input variable in the
formula with the value provided and calculating. Solving will require replacing the
output variable in the formula with the value provided, and solving for the input(s) that
would produce that output.
Chapter 3: Linear Functions 5

Example 5
Given the function k (t )  2t  5
a) Evaluate k(8)
b) Determine t when k(t) = 1

a) To evaluate k(8), we plug in the input value 8 into the formula wherever we see the
input variable t, then simplify
k (8)  2(8)  5
k (8)  16  5
So k(8) = 21

b) To determine t when k(t) = 1, we set the formula for k(t) equal to 1, and solve for
the input value that will produce that output
k(t) = 1 substitute the original formula k (t )  2t  5
2t  5  1 Solve for t.
2t  4
t  2
When solving an equation using formulas, you can check your answer by using your
solution in the original equation to see if your calculated answer is correct.
We want to know if k (t )  1 is true when t  2 .
k (2)  2(2)  5
= 45
= 1 (which was the desired result).

You Try 3.1C

Given the function g(m) = 5m + 17
a. Evaluate g(9)
b. Determine m when g(m) = 2

Graphical Behavior of Functions

As part of exploring how functions change, it is interesting to explore the graphical

behavior of functions.

As we move from left to right (the inputs increase), we say a function is:
 INCREASING if the outputs get larger,
 DECREASING if the outputs get smaller,
 CONSTANT if the outputs do not change.

NOTE: We read graphs just like we read a book…from left to right.

Chapter 3: Linear Functions 6

The following functions are INCREASING

x f(x)
0 4
1 6
2 12
3 24

The following functions are DECREASING

x g(x)
0 10
1 5
2 0
3 -5

The following functions are CONSTANT

x h(x)
0 4
1 4
2 4
3 4

Section 3.2 - Linear Functions

LINEAR FUNCTIONS

A linear function is a function whose graph produces a line.

Linear functions can always be written in the form:

f ( x)  b  mx or f ( x)  mx  b These are equivalent
where b is the initial or starting value of the function (when input, x = 0), and
m is the constant rate of change of the function.

Many people like to write linear functions in the form f ( x)  b  mx because it

corresponds to the way we tend to speak: “The output starts at b and increases at a rate of
m.”
Chapter 3: Linear Functions 7

For this reason alone we will use the form f ( x)  b  mx for many of the examples, but
remember they are equivalent and can be written correctly both ways.

Here, m is the constant rate of change of the function (also called the slope). The slope
determines if the function is an increasing function or a decreasing function.

If m>0 then the function is increasing

If m<0 then the function is decreasing

If m  0 , the rate of change is zero, and the function is a horizontal line

NOTE: A vertical line has a slope that is undefined. A vertical line is NOT a function.

Example 6
Marcus currently owns 200 songs in his iTunes collection. Every month, he adds 15 new
songs. Write a formula for the number of songs, N, in his iTunes collection as a function
of the number of months, m. How many songs will he own in a year?

The initial value for this function is 200, since he currently owns 200 songs, so
N (0)  200 . The number of songs increases by 15 songs per month, so the rate of
change is 15 songs per month. With this information, we can write the formula:
N (m)  200  15m .

N(m) is an increasing linear function.

With this formula we can predict how many songs he will have in 1 year (12 months):
N (12)  200  15(12)  200  180  380 . Marcus will have 380 songs in 12 months.

You Try 3.2A

You have just bought a new Sony 55” 3D television set for $2300. The TV’s value
decreases at a rate of $250 per year. Construct a linear function to represent this
situation. Clearly indicate what your variables represent.

Calculating Rate of Change

Given two values for the input x1 and x2, and two corresponding values for the output, y1
and y2, or a set of points, (x1, y1) and (x2, y2), if we wish to find a linear function that
contains both points we can calculate the rate of change, m:
change in output y y 2  y1
m  
change in input x x2  x1
Chapter 3: Linear Functions 8

Rate of change of a linear function is also called the slope of the line.

Example 7
The population of a city increased from 23,400 to 27,800 between 2002 and 2006. Find
the rate of change of the population during this time span.

The rate of change will relate the change in population to the change in time. The
population increased by 27800 – 23400 = 4400 people over the 4 year time interval. To
find the rate of change, the number of people per year the population changed by:
4400 𝑝𝑒𝑜𝑝𝑙𝑒 1100 𝑝𝑒𝑜𝑝𝑙𝑒
= 1 𝑦𝑒𝑎𝑟 = 1100 people per year
4 𝑦𝑒𝑎𝑟𝑠

(Note: You should recognize the slope as a unit rate from chapter 2!)

Notice that we knew the population was increasing, so we would expect our value for m
to be positive. This is a quick way to check to see if your value is reasonable.

You Try 3.2B

In the year 1998, the surface elevation of Lake Powell was 3,843 feet above sea level. In
the year 2001, the surface elevation of Lake Powell was 3,609 feet above sea level.
Determine the rate of change in this situation.

We can now find the rate of change given two input-output pairs, and can write an
equation for a linear function once we have the rate of change and initial value. If we
have two input-output pairs and they do not include the initial value of the function, then
we will have to solve for it.

Example 8
Working as an insurance salesperson, David earns a base salary and a commission on
each new policy, so David’s weekly income, I, depends on the number of new policies, n,
he sells during the week. Last week he sold 3 new policies, and earned $760 for the
week. The week before, he sold 5 new policies, and earned $920. Find an equation for
I(n), and interpret the meaning of the components of the equation.
The given information gives us two input-output pairs: (3,760) and (5,920). We start by
finding the rate of change.
920  760 160 80
m   (or just 80)
53 2 1
Keeping track of units can help us interpret this quantity. Income increased by $160
$80
when the number of policies increased by 2, so the rate of change is 1 𝑝𝑜𝑙𝑖𝑐𝑦. David earns
a commission of $80 for each policy sold during the week.

We can then solve for the initial value

Chapter 3: Linear Functions 9

I (n)  b  80n then when n = 3, I (3)  760 , giving

760  b  80(3) this allows us to solve for b
b  760  80(3)  520

This value is the starting value for the function. This is David’s income when n = 0,
which means no new policies are sold. We can interpret this as David’s base salary for
the week, which does not depend upon the number of policies sold.

Writing the final equation: I (n)  520  80n

Our final interpretation is: David’s base salary is $520 per week and he earns an
additional $80 commission for each policy sold during the week.

You Try 3.2C

The balance in your college payment account, C, is a function of the number of quarters,
q, you attend. Interpret the function C(q) = 20000 – 4000q in words.

When working with applied problems involving functions, use the following strategy.

Problem Solving Strategy

1) Identify changing quantities, and then carefully and clearly define descriptive
variables to represent those quantities. When appropriate, sketch a picture or define a
coordinate system.
2) Carefully read the problem to identify important information. Look for information
giving values for the variables, or values for parts of the functional model, like slope
and initial value. Also identify what you are trying to find, identify, solve, or
interpret.
3) Identify a solution pathway from the provided information to what we are trying to
find. Often this will involve checking and tracking units, building a table or even
finding a formula for the function being used to model the problem.
4) Solve or evaluate using the formula you found for the desired quantities.
5) Reflect on whether your answer is reasonable for the given situation and whether it
makes sense mathematically.
6) Clearly convey your result using appropriate units, and answer in full sentences when
appropriate.

This strategy will help you work through the following examples.
Chapter 3: Linear Functions 10

Example 9
Arielle saved up $3500 for her summer visit to Seattle. She anticipates spending $400
each week on rent, food, and fun. How long can she afford to stay in Seattle?

In the problem, there are two changing quantities: time and money. The amount of
money she has remaining while on vacation depends on how long she stays. We can
define our variables, including units.

Output: M, money remaining, in dollars Input: t, time, in weeks

Reading the problem, we identify two important values. The first, $3500, is the initial
value for M. The other value appears to be a rate of change – the units of dollars per
week match the units of our output variable divided by our input variable. She is
spending money each week, so you should recognize that the amount of money
remaining is decreasing each week and the slope is negative.

To answer the question, it would be helpful to have an equation modeling this scenario.
Using the intercept and slope provided in the problem, we can write the equation:
M (t )  3500  400t , where M is the amount of money remaining after t weeks.

To find out how long she can stay, we need to find out how long it will take for her to use
up all of the money she has saved. So we are looking for the number of weeks, t, when
M=0. Set the output to zero, and solve for the input:
0  3500  400t
3500
t  8.75
400

Interpreting this, we could say: Arielle will have no money left after 8.75 weeks.

When modeling any real life scenario with functions, there is typically a limited domain
over which that model will be valid – almost no trend continues indefinitely. In this case,
it certainly doesn’t make sense to talk about input values less than zero. It is also likely
that this model is not valid after the horizontal intercept (unless Arielle is going to start
using a credit card and go into debt).

The domain represents the set of input values and so the reasonable domain for this
function is 0  t  8.75 .

However, in a real world scenario, the rental might be weekly or nightly. She may not be
able to stay a partial week and so all options should be considered. Arielle could stay in
Seattle for 0 to 8 full weeks (and a couple of days), but would have to go into debt to stay
9 full weeks, so restricted to whole weeks, a reasonable domain without going in to debt
would be 0  t  8 , or 0  t  9 if she went into debt to finish out the last week.

The range represents the set of output values and she starts with $3500 and ends with $0
Chapter 3: Linear Functions 11

after 8.75 weeks so the corresponding range is 0  M (t )  3500 .

If we limit the rental to whole weeks however, if she left after 8 weeks because she didn’t
have enough to stay for a full 9 weeks, she would have M(8) = 3500 -400(8) = $300
dollars left after 8 weeks, giving a range of 300  M (t )  3500 . If she wanted to stay the
full 9 weeks she would be $100 in debt giving a range of  100  M (t )  3500 .

Most importantly remember that domain and range are tied together, and whatever you
decide is most appropriate for the domain (the independent variable) will dictate the
requirements for the range (the dependent variable).

Example 10
Jamaal is choosing between two moving companies. The first, U-Haul, charges an up-
front fee of $20, then 59 cents per mile. The second, Budget, charges an up-front fee of
$16, then 63 cents per mile1. When will U-Haul be the better choice for Jamal?

The two important quantities in this problem are the cost, and the number of miles that
are driven. Since we have two companies to consider, we will define two functions:
Input: m, miles driven
Outputs: Y(m): cost, in dollars, for renting from U-Haul
B(m): cost, in dollars, for renting from Budget
Reading the problem carefully, it appears that we were given an initial cost and a rate of
change for each company. Since our outputs are measured in dollars but the costs per
mile given in the problem are in cents, we will need to convert these quantities to match
our desired units: $0.59 a mile for U-Haul, and $0.63 a mile for Budget. Looking to
what we’re trying to find, we want to know when U-Haul will be the better choice. Since
all we have to make that decision from is the costs, we are looking for when U-Haul will
cost less, or when Y(m) < B(m).
Using the rates of change and initial charges, we can write the equations:
Y(m) = 20 + 0.59m B(m) = 16 + 0.63m
These graphs are sketched below, with Y(m) drawn dashed.

1
Rates retrieved Aug 2, 2010 from http://www.budgettruck.com and http://www.uhaul.com/
Chapter 3: Linear Functions 12

To find the intersection, we set the equations equal

and solve: Y(m) = B(m).
20 + 0.59m = 16 + 0.63m
-0.04m = -4
m = 100

This tells us that the cost from the two companies

will be the same if 100 miles are driven. Either by
looking at the graph, or noting that Y(m) is growing
at a slower rate, we can conclude that U-Haul will
be the cheaper price when more than 100 miles are
driven.

Example 11
A town’s population has been growing linearly. In 2004 the population was 6,200. By
2009 the population had grown to 8,100. If this trend continues,
a. Predict the population in 2013
b. When will the population reach 15000?

The two changing quantities are the population and time.

Input: t, years since 2004 Output: P(t), the town’s population

The problem gives us two input-output pairs. Converting them to match our defined
variables, the year 2004 would correspond to t = 0, giving the point (0, 6200). Notice
that through our clever choice of variable definition, we have “given” ourselves the
vertical intercept of the function. The year 2009 would correspond to t = 5, giving the
point (5, 8100).

To predict the population in 2013 (t = 9), we would need an equation for the population.
Likewise, to find when the population would reach 15000, we would need to solve for the
input that would provide an output of 15000. Either way, we need an equation. To find
it, we start by calculating the rate of change:
8100  6200 1900 380 people = 380 people per year
m  
50 5 1year
Since we already know the vertical intercept of the line, we can immediately write the
equation: P(t )  6200  380t

To predict the population in 2013, we evaluate our function at t = 9

P(9)  6200  380(9)  9620
If the trend continues, our model predicts a population of 9,620 in 2013.

To find when the population will reach 15,000, we can set P(t) = 15000 and solve for t.
Chapter 3: Linear Functions 13

15000  6200  380t

8800  380t
t  23.158

Our model predicts the population will reach 15,000 in a little more than 23 years after
2004, or somewhere around the year 2027.

You Try 3.2D

In 2004, a school population was 1001. By 2008 the population had grown to 1697.
Assume the population is changing linearly.
a. How much did the population grow between the year 2004 and 2008?
b. How long did it take the population to grow from 1001 students to 1697
students?
c. What is the average population growth per year?
d. Find an equation for the population, P, of the school t years after 2004.
e. Using your equation, predict the population of the school in 2011.

Section 3.3 – Linear Regression

In the last section we explored linear functions. Recall that linear functions have a
constant rate of change, and can be written in the form f(x) = b + mx, or f(x) = mx + b.
When graphed, a linear function is a straight line, where all points that satisfy the linear
equation fall exactly on the graph of the line. This can be seen in the graph below:

This is the graph of f(x)= -1 + x.

Some of the points on this line are: (-

2, -3), (0, -1), (1, 0), and (3, 2). There are
infinitely many other points on this line
too.

Now consider the following graph that represents the relationship between the age of a
vehicle (in years) and its value (in $1000s) for 10 different vehicles.
Chapter 3: Linear Functions 14

Can we write the equation of a linear

function that represents this set of points?
No, not exactly.

There is no straight line that will go

through ALL of these points, but there does
appear to be a linear relationship between
the two variables.

The graph that you see above is called a scatterplot. A scatterplot is used to show the
relationship between two numerical variables. The independent (predictor) variable is
plotted on the horizontal axis, and the dependent (response) variable is plotted on the
vertical axis.

Example 12
The data below show a person’s body weight during a diet program.
Time, t, in weeks 0 1 2 3 4 5
Weight, W, in pounds 196 192 193 190 190 186
Draw a scatterplot of the data. Does the data appear linear?

Step 1: Enter the data into your calculator

 Press STAT (Second Row of Keys)

 Press ENTER to access 1:Edit under EDIT
menu
 Note: Be sure all data columns are cleared.
To do so, use your arrows to scroll up to L1
or L2 then click CLEAR then scroll down.
(DO NOT CLICK DELETE!)

Once your data columns are clear, enter the input data into L1 (press ENTER after each
data value to get to the next row) then right arrow to L2 and enter the output data into L2.
Your result should look like this when you are finished (for L1 and L2):
Chapter 3: Linear Functions 15

Step 2: Turn on your Stat Plot

 Press Y=
 Use your arrow keys to scroll up to Plot1
 Press ENTER
 Scroll down and Plot1 should be highlighted
as at left
 Clear out all entries below

Step 3: Graph the Data in an Appropriate Viewing Window

Click the ZOOM key, and scroll

down to Option 9: “ZoomStat”

 Once your cursor is on “ZoomStat”, press the

GRAPH key. A graph of your data should
appear in an appropriate window so that all
data points are clearly visible.

The scatterplot above shows that

the data points seem to have a
linear relationship.

From the previous scatterplot, it looks like there is a negative linear relationship between
the time in weeks and the weight in pounds. We can see this linear relationship because
the points are scattered around an imaginary straight line. More specifically, as the time
increases, the weight seems to decrease by a fairly consistent amount.
Chapter 3: Linear Functions 16

You Try 3.3A

Consider the data set:

x 3 5 8 9 11 12 15
y 1.1 3.2 7.0 6.8 9.5 9.8 13.0

a. Draw a scatterplot for the data set.

b. Does the data set appear to be increasing or decreasing?
c. Does the data set appear to have a linear relationship?

Just because data are not EXACTLY linear does not mean we cannot write an
approximate linear model for the given data set. In fact, most data in the real world are
NOT exactly linear and all we can do is write models that are close to the given values.

Once we have determined using a scatterplot that our two variables appear to have a
roughly linear relationship, we can use a process called linear regression to find the
equation of the line that best fits the data points. If you take a statistics class, you will
learn a lot more about this process. In this class, you will be introduced to the basics. This
process is also called “FINDING THE LINE OF BEST FIT”. This process is lengthy and
calculation intensive by hand, but can be done quickly using our graphing calulator.

Our calculator will give us the best linear equation possible by taking into account ALL
the given data points.

NOTE: Unless your data are exactly linear, the regression equation will not match all
data points exactly. It is a model used to predict outcomes not provided in the data set.

Example 13
Consider the data set from the previous problem:
Time, t, in weeks 0 1 2 3 4 5
Weight, W, in pounds 196 192 193 190 190 186

Use your graphing calculator to find the regression equation.

Step 1: Enter the Data into your Graphing Calculator

Press STAT then select option 1:Edit under EDIT menu. Clear lists, then enter the
values.
Chapter 3: Linear Functions 17

**NOTE If you ever accidentally DELETE a column, then go to STAT, Option 5:

SetUpEditor>ENTER. When you go back to STAT, your column should be restored.

Step 2: Turn on your Stat Plot and Graph the Data in an Appropriate Viewing
Window
(Refer to previous example for help)

Note: Since we already graphed the scatterplot for this data set (and know the data looks
linear), you could skip this step. Graphing the scatterplot is not required in order to
calculate the regression equation, but you should always verify using a scatterplot that
your data looks linear BEFORE doing a linear regression!

Step 3: Access the Linear Regression section of your calculator

 Press STAT
 Scroll to the right one place to CALC
 Scroll down to 4:LinReg(ax+b)
 Your screen should look as the one at left

Step 4: Determine the linear regression equation

Chapter 3: Linear Functions 18

 Press ENTER twice in a row to view the screen at

left
 The calculator computes values for slope (a) and y-
intercept (b) in what is called the equation of best-
fit for your data.
 Identify these values and round to the appropriate
places. Let’s say 2 decimals in this case.
So, a = -1.69 and b = 195.38
 Now, replace the a and b in y = ax + b with the
rounded values to write the actual equation:
y = -1.69x + 195.38
 To write the equation in terms of initial variables,
we would write W = -1.69t + 195.38
 In function notation, W(t) = -1.69t + 195.38

Once we have the equation figured out, it’s nice to graph it on top of our data to see how
things match up.

GRAPHING THE REGRESSION EQUATION ON TOP OF THE STAT PLOT

 Enter the Regression Equation with rounded

values into Y=
 Press GRAPH
 You can see from the graph that the “best fit” line
does not hit very many of the given data points.
But, it will be the most accurate linear model for
the overall data set.

IMPORTANT NOTE: When you are finished graphing your data, TURN OFF YOUR
PLOT1. Otherwise, you will encounter an INVALID DIMENSION error when trying to
graph other functions. To do this:
 Press Y=
 Use your arrow keys to scroll up to Plot1
 Press ENTER
 Scroll down and Plot1 should be UNhighlighted
Chapter 3: Linear Functions 19

You Try 3.3B

Consider the data set:

n 0 2 4 6 8 10 12
f (n) 23.76 24.78 25.93 26.24 26.93 27.04 27.93

Use your graphing calculator to find the linear regression equation for this data set.

Once you have a linear regression equation, you can then use that equation to provide you
with information about how your two variables change together, and to make predictions
about other values of your variables of interest.

Example 14
Consider again the data set from the previous problem:
Time, t, in weeks 0 1 2 3 4 5
Weight, W, in pounds 196 192 193 190 190 186

a) Interpret the meaning of the vertical intercept.

b) Interpret the meaning of the slope.
c) Use your regression equation to predict the weight in pounds after 2.5 weeks.
d) Use your regression equation to predict the weight in pounds after 6 weeks.

Recall, the regression equation for this data set was

y = -1.69x + 195.38

To write the equation in terms of initial variables,

we would write W = -1.69t + 195.38

a) The vertical intercept is the value of W when t = 0. Based on the regression

equation, we know that when t = 0, W = 195.38. This tells us that when first
starting the weight loss program, the individual weighed approximately 195.38
pounds.

Note: In this particular case, we had a data point for time 0 weeks of 196
pounds. In general, the regression equation gives an estimate (or predicted
value) of the response variable. Output from the regression equation will rarely
equal the exact data values themselves.

b) The slope of the regression equation is -1.69. Recall that the slope measures
Chapter 3: Linear Functions 20

𝑐ℎ𝑎𝑛𝑔𝑒 𝑖𝑛 𝑦 𝑐ℎ𝑎𝑛𝑔𝑒 𝑖𝑛 𝑤𝑒𝑖𝑔ℎ𝑡 (𝑊)

,or in our problem, . We can write the slope for
𝑐ℎ𝑎𝑛𝑔𝑒 𝑖𝑛 𝑥 𝑐ℎ𝑎𝑛𝑔𝑒 𝑖𝑛 𝑡𝑖𝑚𝑒 (𝑡)
this equation of -1.69 as a fraction by writing it over 1:

𝑐ℎ𝑎𝑛𝑔𝑒 𝑖𝑛 𝑤𝑒𝑖𝑔ℎ𝑡 (𝑊) −1.69

=
𝑐ℎ𝑎𝑛𝑔𝑒 𝑖𝑛 𝑡𝑖𝑚𝑒 (𝑡) 1

To interpret the meaning of the slope, we would say that for each additional
week, the weight decreases by approximately 1.69 pounds.

c) To predict the weight in pounds after 2.5 weeks, plug in t = 2.5 to the
regression equation, and solve for W.

W = -1.69(2.5) + 195.38
W = 191 pounds

d) To predict the weight in pounds after 6 weeks, plug in t = 6 to the regression

equation, and solve for W.

W = -1.69(6) + 195.38
W = 185 pounds

In the previous problem, we used our regression equation to tell us how our two variables
change together, and to make predictions about other values. You may have noticed that
we made predictions about the values of our dependent variable only for values of our
independent variable that were close to the actual values in our data set. In other words,
our original data set included values of time from 0-5 weeks. Based on that information,
we created our model. It would not be appropriate to try to make predictions for values
of time that are far outside our original range of time values (from 0 – 5 weeks). This is
called extrapolation and can result in very poor predictions. Consider the following
example.

Example 15
Consider again the data set from the previous problem:
Time, t, in weeks 0 1 2 3 4 5
Weight, W, in pounds 196 192 193 190 190 186

Would it be reasonable to use your regression equation to predict the weight in pounds
after 100 weeks? Why or why not?
Chapter 3: Linear Functions 21

Recall, the regression equation for this data set was

W = -1.69t + 195.38

Our model was created based off of data from 0 to 5 weeks into a weight loss program.
Is it reasonable to assume that the same pattern of weight loss that held in weeks 0-5
would be present in week 100? It certainly doesn’t seem so. This is an example of
extrapolation. Let’s see what would happen if we tried to use the model to predict the
weight in pounds after 100 weeks. We plug in t = 100 to the regression equation, and
solve for W.

W = -1.69(100) + 195.38
W = 26 pounds

If this pattern of weight loss continued, the individual would weigh only 26 pounds
after 100 weeks! This is not likely to be an accurate prediction.

You Try 3.3C

The following table gives the total number of live Christmas trees sold, in millions, in the
United States from 2004 to 2011. (Source: Statista.com).

Year 2004 2006 2008 2010 2011

Total Number of Christmas
Trees Sold in the U.S. 27.10 28.60 28.20 27 30.80
(in millions)

a) Use your calculator to determine the equation of the regression line, C(t) where t
represents the number of years since 2004.

Start by entering new t values for the table below based upon the number of years since
2004. The first few are done for you:

t = number of years since

0 2
2004
C(t) = Total number of
Christmas trees sold in the 27.10 28.60 28.20 27 30.80
U.S. (in millions)

b) Identify the slope of the regression equation and explain its meaning in the context of
this problem.

c) Use the regression equation to predict the number of Christmas trees that will be sold
in the year 2013. Write your answer as a complete sentence.

d) Should you use your regression equation to predict the number of Christmas trees
that will be sold in the year 2030? Why or why not?
Chapter 3: Linear Functions 22

Section 3.4 – Correlation

In the last section we learned about relationships between two variables that appeared to
be linear. In some cases, the scatterplot shows very clearly that the two variables are
linearly related, but it many cases it can be a bit challenging to tell exactly how strong the
linear relationship is between the two variables of interest. The linear correlation (r),
also called the correlation coefficient, is a way to numerically measure the strength and
the direction of the linear relationship between two variables.

Properties of Correlation:

 We use “r” to represent the correlation

 −1 ≤ 𝑟 ≤ 1
 The sign of r indicates the direction of the linear relationship between the 2
variables
o If r < 0, there is a negative linear relationship between the 2 variables
(slope of the regression equation is negative).
o If r > 0, there is a positive linear relationship between the 2 variables
(slope of the regression equation is positive).
 The magnitude of r indicates the strength of the linear relationship between the 2
variables
o If r ≈ 0, then there is no linear relationship between the 2 variables
o If |𝑟| ≈ 1, then there is a very strong linear relationship between the 2
variables
o If r = 1 or -1, then there is an exact linear relationship between the 2
variables (every data point falls exactly on the regression line)

These properties of correlation can be summarized in the graphic below:

Strong, negative Weak Strong, positive

-1 0 1
Chapter 3: Linear Functions 23

Example 16
Consider the following scatterplots, and their corresponding correlation values.

a. b.

r = 0.94 r = -0.15

c. d.

r = -0.88 r = 0.58

Classify the strength and direction of the linear relationship between the 2 variables in
each scatterplot

a. Here r = 0.94, which is close to positive 1. We would say these 2 variables

have a strong, positive linear relationship.
b. Here r = -0.15, which is negative and close to 0. We could say these 2 variables
have a very weak, negative linear relationship.
c. Here r = -0.88, which is close to negative 1. We would say these 2 variables
have a strong, negative linear relationship.
d. Here r = 0.58, which is positive, but not particularly close to 0 or 1. We would
say these 2 variables have a moderate, positive linear relationship.

Note: There is no “cutoff” value for which values of r are considered “strong” and which
are considered “weak”. The interpretation of r can be fairly subjective. What you should
know is that the farther the value of r is from 0, the stronger the linear relationship
between the 2 variables is considered to be.

Now that we have a basic understanding of interpreting the correlation, we can discuss
how to find it. As with linear regression, the calculations for finding correlation by hand
are rather time consuming. We will again rely on the graphing calculator to find the
correlation for us.
Chapter 3: Linear Functions 24

Example 17
Consider again the data set from the weight loss problem:
Time, t, in weeks 0 1 2 3 4 5
Weight, W, in pounds 196 192 193 190 190 186

a. Use your graphing calculator to find the correlation between time and weight.
b. Interpret the meaning of the correlation you found in part a.

To find the correlation on the graphing calculator, press the “2nd” key and then
“Catalog”. “Catalog” is a 2nd function on the zero key. Use the arrow keys to scroll
down to “DiagnosticOn”. Hit ENTER. On the main screen, your calculator should
display DiagnosticOn followed by a blinking cursor. Hit ENTER. The calculator
should display “Done”. Once you have turned on the DiagnosticON, you do not have
to do it again. Also, it will not interfere with any other calculator functions, so you can
just leave it on.

Once you have turned on “DiagnosticOn” you can follow the instructions for finding
the linear regression equation from section 3.3. When the calculator displays the linear
regression equation, it will also display the correlation value, r.

a. Enter your data into L1 and L2, then press STAT, scroll over to CALC, and
select option 4: LinReg(ax+b). Your output should consist of the values of a
and b (just like we did in section 3.3) AND two new values: r2 and r. The
value of r = -0.94. (NOTE: The quantity r2 is the coefficient of determination,
which we do not cover in this course).
b. The value of r = -0.94. This means that time and weight have a strong, negative
linear relationship. This means that as time increases, weight tends to decrease.
This can be confirmed by a scatterplot of the data which showed a negative
slope to the data points.

You Try 3.4

Recall, the following table gives the total number of live Christmas trees sold, in
millions, in the United States from 2004 to 2011. (Source: Statista.com).

t = number of years since

0 2 4 6 7
2004
C(t) = Total Number of
Christmas Trees Sold in the 27.10 28.60 28.20 27 30.80
U.S. (in millions)

a) Use your graphing calculator to find the numerical value of the correlation.

b) Interpret the meaning of the correlation from part a).

Chapter 3: Linear Functions 25

Chapter 3 – Answers to You Try Problems

3.1A a. 51 b. 0

3.1B a. 0 b. 0 or 2

3.1C a. 28 b. -3

3.2A V(n) = 2300 – 250n, where V(n) gives the value (in dollars) of the TV after n
years.

3.2B The surface elevation of Lake Powell is decreasing at a rate of 78 feet per year.

3.2C Your College account starts with $20,000 in it and you withdraw $4,000 each
quarter (or your account contains $20,000 and decreases by $4000 each quarter.)
You can pay for 5 quarters before the money in this account is gone.

3.2D a. 696 students b. 4 years c. 174 students per year d. P(t) = 174t + 1001
e. 2219 students

3.3A a. b. Increasing

c. Yes, there appears to be a linear

trend

3.3B y = 0.32x + 24.16

3.3C a. C(t) = 0.279t + 27.28 b. For each additional year since 2004, the number
of Christmas trees sold in the U.S. will increase by approximately 279,000 trees.
c. 29.79 million Christmas trees
d. No. The original data is for years 2004 through 2011. The year 2030 is too far
from the original data set.

3.4 a. r = 0.52 b. There is a moderate, positive linear relationship between the

time in years since 2004 and the number of Christmas trees sold in the U.S.
Chapter 3 - Practice Problems

1. A function 𝑃(𝑛) gives the wholesale price in dollars of a package of cookies

after n packages have been sold. Interpret the meaning of 𝑃(124) = 0.8.

2. The function 𝐶(𝑔) represents the cost to produce g gallons of chocolate chunk
ice cream. Interpret the meaning of 𝐶(751) = 160.

3. The function 𝐷(𝑡) can be used to approximate the total average credit card debt
in a U.S. household (in thousands of dollars) t years after 1995. Interpret the
meaning of 𝐷(29) = 21.5.

4. A function 𝑑(ℎ) gives the number of miles Billy has driven in his car from his
house h hours after leaving his house.
a. Interpret 𝑑(4) = 160
b. Interpret 𝑑(0.1) = 2

5. The function 𝑔(𝑡) is given by the table below.

t 0 1 2 3 4 5 6
𝑔(𝑡) 12 24 3 24 27 0 3

a. Evaluate 𝑔(1)
b. Evaluate 𝑔(6)
c. Determine t when 𝑔(𝑡) = 24

6. The function 𝑓(𝑥) is represented below as a graph. Use 𝑓(𝑥) to answer the
following questions:

a. Evaluate 𝑓(−1)
b. Evaluate 𝑓(3)
c. Determine x when 𝑓(𝑥) = 0
d. Determine x when 𝑓(𝑥) = 6
Chapter 3: Linear Functions 27

7. The function 𝑓(𝑥) is represented below as a graph. Use 𝑓(𝑥) to answer the
following questions:

a. Evaluate 𝑓(0)
b. Evaluate 𝑓(4)
c. Determine x when 𝑓(𝑥) = −3
d. Determine x when 𝑓(𝑥) = 1

8. Given the function 𝑔(𝑥) = 𝑥 + 2

a. Evaluate 𝑔(−5)
b. Evaluate 𝑔(0)
c. Determine x when 𝑔(𝑥) = −2

9. Given the function 𝑘(𝑡) = −8𝑡 + 9

a. Evaluate 𝑘(3)
b. Evaluate 𝑘(0)
c. Determine t when 𝑘(𝑡) = −39

10. Identify the rate of change and initial value for the function 𝑓(𝑥) = 3𝑥 + 7

11. Identify the rate of change and initial value for the function 𝑓(𝑥) = −9 − 10𝑥

12. Identify the rate of change and initial value for the function 𝑓(𝑥) = −5𝑥

13. Identify the rate of change and initial value for the function 𝑓(𝑥) = 1

14. Calculate the rate of change of the linear function that contains the points (1, 8)
and (4, 17).

15. Calculate the rate of change of the linear function that contains the points
(−15, 30) and (7, −14).

16. Calculate the rate of change of the linear function that contains the points (3, 2)
and (10, 2).
 Chapter 3: Linear Functions 28

17. Determine the rate of change and initial value of the linear function that
generates the following table of values.

x -4 -3 -2 -1 0 1 2 3 4
𝑓(𝑥) -70 -49 -28 -7 14 35 56 77 98

18. Determine the rate of change and initial value of the linear function that
generates the following table of values.

x -4 -3 -2 -1 0 1 2 3 4
𝑓(𝑥) 102 75 48 21 -6 -33 -60 -87 -114

19. Determine the rate of change and initial value of the linear function that
generates the following table of values.

x -4 -2 0 2 4
𝑓(𝑥) -90 -14 62 138 214

20. The function 𝑉(𝑡) = −5.4𝑡 + 34.2 gives the value (in thousands of dollars) of
an investment after t years. Interpret the rate of change in this situation.

21. The function 𝑉(𝑥) = 2.5𝑥 + 25.4 gives the value (in thousands of dollars) of an
investment after x months. Interpret the rate of change in this situation.

22. Paul is planning to sell bottled water at the local carnival. Paul's profit (in
dollars) from selling b bottles of water is given by the formula
𝑃(𝑏) = 1.4𝑏 − 338. Interpret the rate of change in this situation.

23. When a new charter school opened in 1996, there were 440 students enrolled.
Write a formula for the function 𝑁(𝑡), representing the number of students
attending this charter school t years after 1996, assuming that the student
population:
a. Increased by 44 students per year
b. Decreased by 32 students per year
c. Remained constant (did not change)

24. A town's population has been growing linearly. In 2003, the population was
59,000 and the population has been growing by 1,700 people each year.
a. Write a formula for the population x years after 2003.
b. What will the town’s population be in 2007?
c. In what year will the population be 77,700 people?
Chapter 3: Linear Functions 29

25. Last year, Pinwheel Industries introduced a new toy. It cost $1,300 to develop
the toy and $30 to manufacture each toy.
a. Write a formula for the total cost to produce n of these toys.
b. What is the total cost to produce 4,400 toys?
c. How many total toys can be produced with $124,300?

26. In the year 1999, the surface elevation of Lake Powell was 3848 feet above sea
level. In the year 2007, the surface elevation of Lake Powell was 3,227.2 feet
above sea level. Find the rate of change of the surface elevation during this time
span.

27. In the year 1987, an investment was worth $29,800. In the year 1997, this
investment was worth $43,800.
a. Find the rate of change of the investment during this time span.
b. Is the investment increasing or decreasing?

28. In the year 1982, an investment was worth $26,200. In the year 1990, this
investment was worth $39,000.
a. Find the rate of change of the investment during this time span.
b. Is the investment increasing or decreasing?

29. In 1995, the cost of tuition at a large Midwestern university was $170 per credit
hour. In 2000, tuition had risen to $235 per credit hour.
a. Find the rate of change of the cost of tuition during this time span.
b. Write a formula for the cost of tuition as a function of the number of years
since 1990.
c. What will the tuition be in 2007?
d. What year will the tuition be $417 per credit hour?

30. Robert has $100 in his school lunch account. If the school charges $3.50 per
lunch, how many lunches can he purchase? How much money will remain in
his account?

31. You are interested in starting a gym membership and have narrowed your
selection to two gyms. The first, Phoenix Fitness, charges a one-time
membership fee of $99 and charges $35 per month. The second, Slim Gym,
charges a one-time membership fee of $10 and charges $40 per month.
a. Which gym is the more affordable choice if you are going to commit to a
1-year membership?
b. Which gym is the more affordable choice if you are going to commit to a
2-year membership?
Chapter 3: Linear Functions 30

32. a. Use your graphing calculator to create of scatterplot of the data set shown
below.

x 1 3 4 6 7 9 10
y 437 901 1155 1358 1768 2103 2437

b. Does the data set appear to be increasing or decreasing?

c. Does the data set appear to be roughly linear?

33. a. Use your graphing calculator to create of scatterplot of the data set shown
below.

x 2 9 14 23 33 42
y -60.2 -130.1 -243.7 -328.9 -580.5 -643.8

b. Does the data set appear to be increasing or decreasing?

c. Does the data set appear to be roughly linear?

34. a. Use your graphing calculator to create of scatterplot of the data set shown
below.

x 8 12 14 23 24 29 33
y 10 20 15 24 35 30 28

b. Does the data set appear to be increasing or decreasing?

c. Does the data set appear to be roughly linear?

35. a. Use your graphing calculator to create of scatterplot of the data set shown
below.

x 20 24 27 32 48 60 68
y 12 20 28 41 24 15 11

b. Does the data set appear to be increasing or decreasing?

c. Does the data set appear to be roughly linear?

36. Calculate the linear regression equation for the data from problem 32:
x 1 3 4 6 7 9 10
y 437 901 1155 1358 1768 2103 2437

37. Calculate the linear regression equation for the data from problem 33:
x 2 9 14 23 33 42
y -60.2 -130.1 -243.7 -328.9 -580.5 -643.8
Chapter 3: Linear Functions 31

38. Calculate the linear regression equation for the data from problem 34:
x 8 12 14 23 24 29 33
y 10 20 15 24 35 30 28

39. Consider the following 4 scatterplots. Match each scatterplot with the most
appropriate value from the 4 correlations listed below.

a. b. b.

r = ________ r = ________

c. d. d.

r = _________ r = _________

r = 0.38 r = 0.94 r = -0.15 r = -0.85

40. Find and interpret the linear correlation for the data from problem 32:

x 1 3 4 6 7 9 10
y 437 901 1155 1358 1768 2103 2437

41. Find and interpret the linear correlation for the data from problem 33:

x 2 9 14 23 33 42
y -60.2 -130.1 -243.7 -328.9 -580.5 -643.8

42. Find and interpret the linear correlation for the data from problem 34:

x 8 12 14 23 24 29 33
y 10 20 15 24 35 30 28
Chapter 3: Linear Functions 32

43. Find and interpret the linear correlation for the data from problem 35:

x 20 24 27 32 48 60 68
y 12 20 28 41 24 15 11

44. The following table shows the number of newspaper subscriptions in

Middletown, USA where t represents the number of years since 2002 (t = 0 in
2002) and S(t) represents the total subscriptions each year measured in
thousands.

t (year) 0 2 4 6 8
S(t) (total subscriptions in 448 372 198 145 45
1000’s)

Find and interpret the linear correlation.

45. Scott is hiking the Appalachian Trail from Georgia to Maine. The distance of
his hike is 2200 miles. It took Scott 123 days to complete the hike. The data
below represent the distance, D, he had hiked t days after the start of his trip.

t (days hiking) 0 32 47 73 99 123

D(t) (distance in 0 590 912 1212 1876 2200
miles)

Find and interpret the linear correlation.

46. Which implies a stronger linear relationship, a correlation of -0.84 or -0.47?

Explain.

47. Which implies a stronger linear relationship, a correlation of -0.73 or 0.51?

Explain.

48. The following table shows the number of newspaper subscriptions in

Middletown, USA where t represents the number of years since 2002 (t = 0 in
2002) and S(t) represents the total subscriptions each year measured in
thousands.

t (year) 0 2 4 6 8
S(t) (total subscriptions in 448 372 198 145 45
1000’s)

a) Use your graphing calculator to create of scatterplot of the data set.

b) Based on your graph above, do the data appear to be approximately linear?

Chapter 3: Linear Functions 33

c) Use your calculator to determine the regression equation in S(t) = at + b form.

(Round to 2 decimal places)

d) What is the slope of your regression model for S(t) and what is its meaning in the
context of this problem?

e) What is the initial value of your linear regression model for S(t) and what is its
meaning in the context of the problem.

f) Use your linear regression equation to estimate the total number of subscriptions
in 2007 (i.e. when t =5). Show your computations here and your final result.

g) Use your linear regression equation to estimate the total number of subscriptions
in 2004. How does this value compare to the data value in the table?

h) Should you use your linear regression equation to estimate the circulation in the
year 2030? Why or why not?

49. Scott is hiking the Appalachian Trail from Georgia to Maine. The distance of
his hike is 2200 miles. It took Scott 123 days to complete the hike. The data
below represent the distance, D, he had hiked t days after the start of his trip.

t (days hiking) 0 32 47 73 99 123

D(t) (distance in 0 590 912 1212 1876 2200
miles)

a) Use your graphing calculator to create of scatterplot of the data set.

b) Based on your graph above, do the data appear to be approximately linear?

c) Use your calculator to determine the regression equation in D(t) = at + b form.

(Round to 2 decimal places)

d) What is the slope of your regression model for D(t) and what is its meaning in the
context of this problem?

e) Use your linear regression equation to estimate the total number of miles Scott
has hiked in 50 days. Show your computations here and your final result.

f) Should you use your linear regression equation to estimate when Scott has hiked
2500 miles? Why or why not?