Module 7: Probability

Module 7: Probability

7.01 Learning Objectives

Module 7: Learning Objectives

After completing this module, you should be able to:

1. Relate the probability of an event to the likelihood of the event occurring

2. Explain how relative frequency can be used to estimate the probability of an outcome or an event
3. Determine the sample space of a given random experiment
4. Find the probability of events in the case in which all outcomes are equally likely
5. Determine if two events are disjoint
6. Apply the addition rule to find the probability of a given set of disjoint events
7. Determine if two events are independent
8. Apply the multiplication rule to find the probability of a given set of independent events
9. Apply basic probability concepts to interpret and solve mathematical problems
10. Find probabilities using Venn diagrams or probability tables as aids
11. Explain the concept of conditional probability and properly use the probability notation for a conditional event
12. Identify an example that represents a case of conditional probability
13. Determine the probability of an event using principles of conditional probability

7.02 Probability
Introduction to Probability

Copyright © 2019 MindEdge Inc. All rights reserved. Duplication prohibited.

What is probability? Put simply, it is the chance of an event occurring. Probability is often expressed in quantifiable terms as
either a percentage or a decimal: for example, the chance of a coin landing as heads is 50 % . It is impossible to be certain you
can correctly predict whether the coin will land heads or tails, which makes the act of flipping the coin a random event. We can
however, calculate the probability of the coin landing heads or tails.

What is Probability?

Slide 1

One familiar use of probability is the weather.

Slide 2

If you are hosting a barbecue in the summer, you might check ahead to see the probability it will rain on the
Wednesday you have planned your party.

Slide 3

Copyright © 2019 MindEdge Inc. All rights reserved. Duplication prohibited.

 If the probability is 90 % , you might be tempted to change the party date—or at least get ready to host it inside.

Slide 4

Because they are about uncertain future events, probabilities are different from another common type of information
about the future. If we say to a friend "I'll meet you for lunch tomorrow" and then fail to show up, there is a problem.
We have done something wrong or something bad has happened to us. The information we gave our friend was
supposed to be a certainty.

Slide 5

Copyright © 2019 MindEdge Inc. All rights reserved. Duplication prohibited.

 In contrast, if it does not rain on the planned day of our party, it does not mean the probability of90 % chance of rain
was "wrong." It might mean that we were lucky and had the less likely—but still possible—scenario of no rain. That is,
you were in the lucky 10 % of cases where those weather conditions did not lead to rain. A probability less than
100 % is never a certainty; it just tells us the likelihood of an occurrence.

Slide 6

More specifically, a probability, such as 90 % , tells us that when weather conditions are the same as they are on
Wednesday,
9 times out of
10 it will rain. If the conditions were exactly reproduced 100 times, about 90 of those times, it would rain and about 10
of those times it would not rain.

Probability and Subjectivity

Probability is a subjective area that often times requires judgement. In most all cases the extremes (impossible, certain) hold
true. However in other cases, what one person might say is "unlikely," another person may find is "as unlikely as it is likely."
Therefore, it is important to be mindful that while there is a measure of objectivity to many situations where probability is
concerned, there are other situations where the likelihood of an outcome is often defined in the eye of the beholder.

Examples of Quantitative and Qualitative Probability

Qualitative
Quantitative Probability Examples
Description
Impossible 0 % probability Probability that the day of the month will be32
More than 0 % but less than 40 %
Unlikely A snowstorm in Boston leaves more than15 inches of snow
probability
As likely as unlikely 40 % to 60 % probability A pregnant woman having a boy rather than a girl
More than 60 % but less than 100 % The probability that a child born in the United States will live to
Likely
probability adulthood
Certain 100 % probability The probability that everyone alive today has a birthday

Career Connections

Probability

Since many important decisions have to be made in the absence of knowledge about the future, probability is a crucial
business tool. Decisions about staffing, capital purchases, and inventory levels (to name just three key components of a
company) need to be made before the company knows exactly what demand there will be for its goods or services. Probability
puts hard numbers in front of managers enabling them to make better decisions.

Copyright © 2019 MindEdge Inc. All rights reserved. Duplication prohibited.

Probability can help by offering the company the likelihood of different scenarios. A manager can
run an analysis that considers what will happen to sales under the best-case scenario, the most
likely scenario, and the worst-case scenario. Each scenario has a probability attached to it. So
sales holding steady might have a probability of 40%, sales increasing by up to 10% might have a
probability of 20%, sales increasing by 10-20% might have a probability of 5%, etc. Using this
information, the manager can project future staffing needs under each scenario and make a more
informed decision about hiring or firing personnel.

Suppose someone in management in your company has years of experience. Couldn't he just
estimate these probabilities in his head? Yes, but using probability allows him to handle additional complexity. There is also a
set of probabilities for how many employees will retire within the next six months and how many will leave the company.
Feeding these numbers into an algorithm will yield a set of different scenarios that give a more complete picture of whether
staffing is likely to fall short or exceed what the company needs.

Decimal Expression of Percentage

Probability percentages range strictly from 0 % to 100 % . It makes no sense mathematically to say there is110 % chance that X
will occur—although people often say such things to emphasize their point. Therefore the decimal form of percentages range
from 0 to 1 (remember that to obtain the decimal form of a percentage, you divide by100). A 0 probability means the event will
definitely not occur, while 1 probability means the event will definitely occur. As the decimal values get closer to1, the
probability of the event occurring increases. An event with a probability of 0. 35 (35 % ) is unlikely; an event with a probability of
0. 65 (65 % ) is likely; and an event with a probability of0. 9 (90 % ) is very likely.

Many probabilities are calculated by creating a fraction (as we will learn on the next page). Therefore to calculate and express
probabilities, you will need to be comfortable with percentages, decimals, and fractions and know how to convert among them.
For a review of this topic, please see page 2.14.

Before we go on, it is important to understand the mathematical language and notation around probability

Expressing Probability

Probability, like other topics in math, brings with it a specialized vocabulary. First, read the quick example before the formal
definitions are given:

Two friends bet on a roll of a die. If an even number,2, 4, or 6, appears, Jayne wins, while if an odd number,1, 3, or 5,
appears, Stan wins. The chance of either person winning is 50 % . Mathematicians and statisticians would call the roll of the
die an "experiment." Each possible number that the die can land on is known as an "outcome." The possible outcomes of
this experiment are rolling a 1, 2, 3, 4, 5, or 6. An "event" is one or more outcomes of an experiment. The die landing on an
even number is an event, as is the die landing on an odd number. If all of the outcomes are equally likely, a statistician
would call the die (or the experiment) "fair."

An experiment is the procedure, or situation, for which the probability is being calculated. So the roll of a die can be an
experiment, a coin flip can be an experiment, or even the weather on a particular Wednesday. The possibilities of what can
occur during an experiment—the results of the experiment—are called outcomes. When rolling a die,1, 2, 3, 4, 5, and 6 are
the possible outcomes. An event is comprised of one or more outcomes. The die landing on an even number is an event.
The die landing on 1 is also an event. The probability of an event is represented byP(E)
which means probability (P
) of a certain event (E
) occurring. If R =
Rain and the weatherman says there is a 40
percent chance of rain, then P(R) = 0. 40
. (Notice that the probability is written in its decimal form). An experiment is fair if each outcome is equally likely.

Career Connections

Machine Learning

Have you ever heard of "machine learning?" Machine learning is when computers learn or respond to situations differently
based on previous data. If you've ever played with the popular app "Akinator," you've played with a program/app that is a great

Copyright © 2019 MindEdge Inc. All rights reserved. Duplication prohibited.

example of machine learning. Another great example is the predictive text feature in your cell
phone. You type in "Elep" and your phone predicts that you might want "Keep" or "Elephants." But
how does machine learning work? Through tracking probabilities! By tracking probabilities of
certain events, computer programs can predict the most likely outcome to a certain event.

Facebook and Google also use machine learning to target specific ads to you; they basically offer
you ads that you are most likely to be interested in. Again, all of this works by looking at an
individual user and figuring out what that particular user is most likely to respond to, to be
interested in (in terms of an ad), or even in predicting the next word in a text message. None of
these key technology features would be possible without probabilities.

7.03 Sets
Sets
Throughout this course, we've been working with datasets. We've also seen sets as they relate to basic numeracy (the set of
integers, the set of prime numbers, etc.). Formally, a set is simply a collection of unique elements.

For example, a set of tree species is: oak, juniper, elm, maple.

Notation

As a reminder, sets are often listed between brackets, with commas separating each element in the set:

{
oak, juniper, elm, maple}

In mathematics, a set is often, but not always, a collection of numbers:

{ - 38, - 1, 1, 35, 63, 94}

We often represent sets with a symbol or a letter, much like variables. For example, we can label a set asA:

A = { - 38, - 1, 1, 35, 63, 94}

By representing this set with a letter, we can more easily refer to this set. Rather than listing all of the elements every time the
set is referenced, we can simply say, "A
has six elements" or "every number in A
is an integer."

Subsets

In math and statistics, we often work with multiple sets at once. These sets can relate to one another. One such relation is
known as a subset.

Set A
is a subset of set B
, if every element in A
is contained within B
.

For example:

A = {1, 2, 3}

B = {1, 2, 3, 4, 5}

A
is a subset of B
, because every element in setA
is contained within set B
.

Copyright © 2019 MindEdge Inc. All rights reserved. Duplication prohibited.

Empty Set

An empty set is a set that has no elements. There is nothing in the set; therefore, it is empty. This may seem odd, or even like
it isn't a set at all, but an empty set is, in fact, a set.

For example, let's say you wanted to list the days of the week that do not end in a y. There are none! Therefore, this is the
empty set. In set notation, the empty set is written as a pair of brackets with nothing between them:

{}

Union

The union of two sets is a collection of all of the elements listed in the sets.

For example:

C = {2, 4, 6}

D = {1, 3, 5}

The union of C
and D
is {1, 2, 3, 4, 5, 6}
, as those are all of the elements that appear in the sets.

Intersection

The intersection of two sets is a collection of the elements listed in both of the sets.

For example:

E = {0, 10, 100}

F = { - 2, - 1, 0, 1, 2}

The intersection of E
and F
is {0}
, as 0
is the only element that appears in both sets.

7.04 Probability and Relative Frequency

Probability and Relative Frequency

Theoretical (Classical) Probability

An intuitive mathematical formula gives us a number that measures probability. If we are betting that a flipped coin will land on
heads, we know there are two possibilities, heads and tails, and we win with only one of those possibilities (heads showing). It
seems reasonable that the probability we will win this bet is 50 %
1
or 2
if expressed as a fraction.

Theoretical probability is calculated as the number of ways one particular event can occur in a random experiment, divided by
the total number of possible outcomes:

Copyright © 2019 MindEdge Inc. All rights reserved. Duplication prohibited.

For example, Bob has 5
different suits:

If Bob likes the suits equally—that is, each morning he chooses one randomly—what is the probability he will choose to wear
blue?

P(Bob wearing a blue suit)

=
3 blue suits
5 suits total
3
= 5
3
The probability that Bob wears a blue suit is 5
. This can also be expressed as 0. 6
or 60 %
.

Example
In a marketing study, there are four different experimental groups categorized as Group A, Group B, Group C, and Group D. If
a participant is equally likely to be assigned to any group, what is the probability he or she is assigned to Group D?
There are a total of 4
groups.

( )
Number of Group D 1
P assigned to Group D = Total number of groups = 4

Copyright © 2019 MindEdge Inc. All rights reserved. Duplication prohibited.

 1
Note that 4
can be expressed as 0. 25
or 25 %

Empirical (Observational) Probability

To calculate theoretical probability, each outcome must be equally likely. The problem is that often in real life, outcomes are
not equally likely.

The classic example of a situation where theoretical probability does not provide much insight is when calculating the
probability that a sports team will win a certain game. Theoretical probability would have to assume that each team is equally
likely to win. But we know that in real life, often one team is far more likely to win than the other. We want a way to incorporate
what we know about the teams to get a better sense of their chances.

In this case, mathematicians turn to empirical probability to estimate the probability of an event. Empirical probability gathers
data by performing multiple experiments, or trials, and recording the results each time. For example, to have a better idea of
the probability that a team will win a particular game, we naturally examine what other games it has won and against which
teams. Empirical probability provides an estimate of the theoretical probability of an event from the relative frequency, i.e how
often the event occurs in the series of trials (or experiments) relative to the number of trials.

Accurately measuring relative frequency depends on having random experiments. Random experiments are trials in which the
outcome is not known ahead of time and the result does not depend on the results of other trials. For example, you can flip a
coin ten times, and each will be a random experiment. You do not know ahead of time whether you will get heads or tails.
Getting heads or tails on one flip does not help you predict the result of the next flip.

Calculating Relative Frequency

Empirical probability is estimated using the actual results of experiments or trials.

Note that relative frequency only gives an estimate of the probability of an event occurring. As any sports fan knows,
even if the same team (same players, same coach, etc.) played in the exact same league (same players, same
coaches, same weather, etc.), their results two seasons in a row are unlikely to be the same. That is, for any two sets
of experiments, the results vary each time.

The first season this hypothetical team might win 25

out of 30
games; the next season, the team might only win 23
out of 30
25
games. Is the "true" probability that this team will win the next game 30
23
or 30
?

Copyright © 2019 MindEdge Inc. All rights reserved. Duplication prohibited.

 Either number is an estimate of the probability it will win the next game, but either number is more accurate than a
50 %
probability of winning that theoretical probability would give.

Relationship between Theoretical and Empirical Probabilities

Relationship between Theoretical and Empirical Probabilities

We can discover the relationship between the theoretical and empirical probabilities by running experiments for which
the theoretical probability is accurate, such as flipping a coin. The probability that someone will flip a "heads" in a coin
toss is the number of desired outcomes (1
, the head side) divided by the total number of possibilities (2
1
, heads or tails). 2
is 0. 5
or 50 %

Example and Exercise

How does 50 %
compare to the relative frequency of heads in a set of data? See for yourself. The following Interactive Coin Toss
activity will be used to answer the relative frequency questions below. Read the instructions and then click on the link
below to run a couple of sets of experiments to answer the questions.

First run an experiment by flipping a coin 10

times. Click on "No" for the question "Run Cumulative Stats" since we want each experiment to be isolated from the
other experiments.

Insert "10
" into the "number of tosses" box. Click "Toss 'em" to run the experiment. To get the relative frequency, click "Ratio"
on the side of the box to see the number of heads as a fraction.

Now, run the experiment for 10

coin tosses, next for 100
coin tosses, and finally for 1000
coin tosses. For each trial record the decimal answer in the white box next to the corresponding question below."

Shodor Interactive Coin Toss

Copyright © 2019 MindEdge Inc. All rights reserved. Duplication prohibited.

What do you notice about the relationship between the experimental data and the theoretical probability?

1. What was the relative frequency of heads for10

coin tosses?

Click for feedback

The relative frequency will be between 0

and 1. 0
. We expect these values to range widely; maybe you got as many as7
or as few as 3
heads in this experiment.

2. What was the relative frequency of heads for100

coin tosses?

Click for feedback

The relative frequency will be between 0. 3

and 0. 7
. We expect these values to range more narrowly. Probability is not certainty, so you may toss 0
heads or 100
, but most likely your relative frequency was closer to0. 5
.

3. What was the relative frequency of heads for 1000 coin tosses?

Click for feedback

The relative frequency will be between 0. 4

and 0. 6

Copyright © 2019 MindEdge Inc. All rights reserved. Duplication prohibited.

 . We expect these values to range even more narrowly than the last set. Most likely, your relative frequency was
between 48 %
to 52 %
.

The Law of Large Numbers

The Law of Large Numbers

The law of large numbers states that as the number of trials increases, the relative frequency of an event will
converge on the theoretical probability. Simply put, the more times you flip a coin, the closer the relative frequency of
heads will be to 50 %
.

Sample Relative Frequencies of Tossing Heads

Trials Number of heads Relative Frequency of Heads Distance from Expected Value of 0. 5
10 7
0. 7 0. 2
flips of a coin heads
100
47 0. 47 0. 03
flips of a coin
1000
492 0. 492 0. 008
flips of a coin
10000
4997 0. 4997 0. 0003
flips of a coin
100000
49884 0. 49884 0. 00116
flips of a coin

Notice a couple of things about the trials above. First, each of the results, except the first, had a relative frequency
less than the expected 0. 5
, but that is just random. Other trials would result in relative frequencies greater than the expected value. Second,
although the difference between the relative frequencies and the expected value generally decreased, it did not do so
uniformly. Again, each trial is random and the results cannot be guaranteed. Nevertheless it is true that as the number
of flips gets larger, the relative frequency will get closer to 0. 5
.

7.05 Sample Space

Sample Space
Calculating theoretical probability requires knowing the total number of (equally likely) possible outcomes. This set of possible
outcomes in an experiment is called the sample space. The sample space of flipping a coin is heads and tails. The sample
space of rolling a regular six-sided die is 1
,2
,3
,4
,5
, and 6
. While the sample space is all the different outcomes, the sample size is the number of different outcomes. We need the
sample space to determine the sample size, and we need the sample size to calculate probability. The sample size of flipping
a coin is 2
; the sample size of rolling a six-sided die is6
.

Finding the sample space for flipping a coin or rolling a die is very straightforward, but some probability questions have more
complicated sample spaces. There are a couple of techniques for determining sample space and sample size.

Copyright © 2019 MindEdge Inc. All rights reserved. Duplication prohibited.

A List

One very straightforward way to determine a sample space is to write out all the options in a list. The downside of this
method is it can be difficult to determine if you have missed an outcome. To show the method, suppose someone who
is planning to have two children wants to know the probability of having one boy and one girl. To calculate the
probability, you first must determine the sample space of possible outcomes. This is a relatively easy sample space to
find.

Sample Space of Having 2 Children

Boy Boy
Boy Girl
Girl Boy
Girl Girl

You can see from the table above that the sample size is4
and there are two ways to have a boy and a girl. The probability of having 1
boy and 1
girl when having 2
2 1
children is 4 = 2
.

When making a list, be sure to create as much of a pattern as you can to ensure you have exhausted all possibilities.
For example, you might organize the list above by starting with the maximum number of boys you can get.

Now, let us try a more complicated example. Try to list all the possible ways to have three children in the box below.
Make each item in the list its own line.

Exercise: Sample Space of Having 3 Children

Without a lot of organization and concentration, it is easy to miss one outcome in the sample space. Other methods for
discovering a sample space are more reliable.

A Tree

A tree is a way of writing a list that expands horizontally rather than vertically. It ensures that you do not miss an
outcome. To create a tree to help you determine the sample space, start by writing the first set of options. Attach the
next set of options to each item in the first set.

Copyright © 2019 MindEdge Inc. All rights reserved. Duplication prohibited.

Each time you add options, make sure you include all of them for each of the earlier options.

Copyright © 2019 MindEdge Inc. All rights reserved. Duplication prohibited.

You will notice that as you add more options, you will need to include them more times. The final row of "boy/girl" is
added four times.

Once you have all the outcomes in the sample space listed, you read across the tree from left to right to get one
unique outcome. The first "row" of this tree is boy-boy-boy, stretching upward and to the right. The second row is boy-
boy-girl. You can also easily count the number of unique outcomes. Since each branch represents a unique outcome
in the sample space, you count the last items.

Copyright © 2019 MindEdge Inc. All rights reserved. Duplication prohibited.

 The sample size of "different ways of having three children" is8

Because the tree is systematic, you can be sure not to have missed an outcome.

A Table

The tree approach to finding the sample space also has limitations. It can become difficult to squeeze in every option if
there are more than two. For more than two options, it is best to use a table.

For a table, label the top and sides with the different options. Fill the cells in with the combination of options. The
following is an example of a table illustrating the outcomes of a rolling two six sided die.

Copyright © 2019 MindEdge Inc. All rights reserved. Duplication prohibited.

 The table again will help you list out every possible outcome in the sample space by organizing the outcomes of the
two die rolls.

Something Must Happen

It is important to remember that the probabilities of all the outcomes in the sample space always sum to
1. That is, there is a 100
percent chance of one of the outcomes in the sample space happening, since the sample space, by definition, contains all
possible outcomes.

Tree Exercise 1

Tree Exercise 2

7.05.1 Sample Spaces and Probability

Sample Spaces and Probabilities

In the previous example of having three children, we should assume that having a boy is as likely as having a girl (though in
practice having a boy is slightly more likely). Using our basic rule of probability, we find the likelihood by calculating the number
of ways for the desired event to occur divided by the number of total outcomes. We can use the tree to calculate different
probabilities. For the following probabilities, we are assuming a family has or is planning to have exactly three
children.

Copyright © 2019 MindEdge Inc. All rights reserved. Duplication prohibited.

 Example Probabilities from the Children Tree
Question Explanation Probability
This is one possibility out of the total of8
different options.

What is the probability that a family will 1

have a boy and then a girl and then another
8
boy?

There are 2
ways to have children of the same gender: 3
boys or 3
girls

Copyright © 2019 MindEdge Inc. All rights reserved. Duplication prohibited.

What is the probability that a family will 2 1
have 3 8 = 4
children of the same gender?

To do this problem, count up the number of tree branches that

contain at least one boy. That is every row except girl-girl-girl, so a
total of 7
.

What is the probability that a family will 7

have at least one boy? 8

Copyright © 2019 MindEdge Inc. All rights reserved. Duplication prohibited.

 Again, count the number of rows that contain 2
girls and divide this number by 8
.
The rows are boy-girl-girl
girl-boy-girl
and girl-girl-boy

What is the probability that the family will 3

have exactly 2
8
girls?

Calculating probabilities in which each outcome is equally likely can have limited uses in business or IT. Many more important
probabilities are calculated when the outcomes are not equally likely. We will learn how to do this later in the module.

Exercises

Copyright © 2019 MindEdge Inc. All rights reserved. Duplication prohibited.

 Exercise: Sample Spaces and Probabilities
The following tree graph will be used for Questions 1
through 4
. Use the / sign to indicate division. Always reduce fractions to lowest terms

1. What is the probability of a family that has two girls with a boy in between?
2. What is the probability that a family has at least two children of the same gender?
3. What is the probability that a family has exactly one boy?
4. What is the probability that a family does not have two children of the same gender in a row?
The following tree graph will be used for Questions 5
through 7
.

TechTown is analyzing the purchasing behavior of their customers in order to gain insight as to how they can better manage
their inventory. In order to calculate the quantity of each product they should have on hand, we first need to calculate the
probability of a customer purchasing of product. As TechTown's purchasing manager, you have been tasked with calculating
the probabilities. Assume the probability of each outcome is equally likely. Use the / sign to indicate division. Always
reduce fractions to lowest terms

Copyright © 2019 MindEdge Inc. All rights reserved. Duplication prohibited.

5. What is the probability that a customer purchased a laptop (assuming all outcomes are equally likely)?
6. What is the probability that a customer purchased a model with a display resolution of1920x1080
p?
7. What is the probability that a customer purchased a desktop with a22
inch monitor?
The following table will be used to answer Questions 8
through 10
. Enter the result as a fraction. Use the / sign to indicate division. Always reduce fractions to lowest terms.

Number 1 2 3 4 5 6 7 8 9 10
1 2 3 4 5 6 7 8 9 10
Heads
,H ,H ,H ,H ,H ,H ,H ,H ,H ,H
1 2 3 4 5 6 7 8 9 10
Tails
,T ,T ,T ,T ,T ,T ,T ,T ,T ,T
8. What is the probability of an even number and tails?
9. What is the probability of a number greater than or equal to8
?
10. What is the probability of getting heads?

7.06 Category Concepts

Category Concepts
Throughout this module, we have been primarily examining events comprised of one outcome. Now we will broaden our scope
to include more complicated events.

Consider a man named Bob who has five suits and three pairs of shoes. He has one black suit, one brown suit, and three blue
suits. He has two pairs of black shoes and one pair of brown shoes. We can put the two categories of clothes together. The
table below lays out all the different possibilities of shoes and suits.

Copyright © 2019 MindEdge Inc. All rights reserved. Duplication prohibited.

 Bob's black
black shoes Outfitsshoes brown
1 2 shoes
black suit, black suit, black suit,
black suit
black shoes 1 black shoes 2 brown shoes
brown brown suit, brown suit, brown suit,
suit black shoes 1 black shoes 2 brown shoes
blue suit 1 blue suit 1 blue suit 1
blue suit 1 , , ,
black shoes 1 black shoes 2 brown shoes
blue suit 2 blue suit 2 blue suit 2
blue suit 2 , , ,
black shoes 1 black shoes 2 brown shoes
blue suit 3 blue suit 3 blue suit 3
blue suit 3 , , ,
black shoes 1 black shoes 2 brown shoes

Let A
=
Bob wears a black suit. Let B
=
Bob wears black shoes. We can now use this table to define some important categories in probability.

Universe

The universe is the entire category being considered. It encompasses bothA

and B
and outcomes that are not in either category.

In our example of Bob's shoes and suits, the universe would be Bob's outfits, the ensemble of suits and shoes. These
outfits would not include shirts, ties, or socks! Every cell in the table would be included in the universe.

"Not"

The word "not" before a category is everything outside that category but within the universe.

In the table below, "not A" is illustrated in orange. RememberA

=
"Bob wears a black suit." "Not A
" is Bob wears a blue or brown suit. The color of the shoes is irrelevant.

Bob's Outfits
black shoes 1 black shoes 2 brown shoes
black suit, black shoes black suit, black shoes
black suit black suit, brown shoes
1 2
brown suit, brown suit,
brown brown suit, brown
black shoes black shoes
suit shoes
1 2
blue suit blue suit
blue suit
blue suit 1, 1,
1,
1 black shoes black shoes
brown shoes
1 2
blue suit blue suit
blue suit
blue suit 2, 2,
2,
2 black shoes black shoes
brown shoes
1 2
blue suit blue suit
blue suit
blue suit 3, 3,
3,
3 black shoes black shoes
brown shoes
1 2

Copyright © 2019 MindEdge Inc. All rights reserved. Duplication prohibited.

 Not B
refers to outcomes that are not in the category of B but still in the universe.

In our Bob example, "not B

" is not wearing black shoes. This is again illustrated in orange in the table below.

Bob's Outfits
black shoes black shoes brown
1 2 shoes
black suit, black suit, black suit,
black suit
black shoes 1 black shoes 2 brown shoes
brown brown suit, brown suit, brown suit,
suit black shoes 1 black shoes 2 brown shoes
blue suit 1 blue suit 1 blue suit
blue suit 1 , , 1,
black shoes 1 black shoes 2 brown shoes

blue suit 2 blue suit 2 blue suit

blue suit 2 , , 2,
black shoes 1 black shoes 2 brown shoes
blue suit 3 blue suit 3 blue suit
blue suit 3 , , 3,
black shoes 1 black shoes 2 brown shoes
A
and B
(intersection)

A and B signals the set of elements where both A and B are true. It is called the intersection of A and B.

For Bob's outfits, A and B would be the situations in which Bob is wearing both a black suit and black shoes. A and B
is represented below with orange cells.

Bob's Outfits
black shoes black shoes brown
1 2 shoes
black suit, black suit,
black suit,
black suit black shoes black shoes
brown shoes
1 2
brown brown suit, brown suit, brown suit,
suit black shoes 1 black shoes 2 brown shoes
blue suit 1 blue suit 1 blue suit 1
blue suit 1 , , ,
black shoes 1 black shoes 2 brown shoes
blue suit 2 blue suit 2 blue suit 2
blue suit 2 , , ,
black shoes 1 black shoes 2 brown shoes
blue suit 3 blue suit 3 blue suit 3
blue suit 3 , , ,
black shoes 1 black shoes 2 brown shoes
A
or B
(union)

A
or B
is the category which includes all the outcomes in A
, all the outcomes in B
and all the outcomes in their intersection. It is called the union of A
or B
.

Copyright © 2019 MindEdge Inc. All rights reserved. Duplication prohibited.

 For Bob's outfits, A
or B
would be the situations in which Bob is wearing a black suit, black shoes, or both. A
or B
is represented below with orange cells.

Bob's Outfits
black shoes black shoes brown
1 2 shoes
black suit, black suit,
black suit,
black suit black shoes black shoes
brown shoes
1 2
brown suit, brown suit,
brown brown suit,
black shoes black shoes
suit brown shoes
1 2
blue suit blue suit
blue suit 1
1, 1,
blue suit 1 ,
black shoes black shoes
brown shoes
1 2
blue suit blue suit
blue suit 2
2, 2,
blue suit 2 ,
black shoes black shoes
brown shoes
1 2
blue suit blue suit
blue suit 3
3, 3,
blue suit 3 ,
black shoes black shoes
brown shoes
1 2

Venn Diagrams

Now let us consider the same categories using a more real-world example and a different visual aid: the Venn diagram that
defines different categories spatially.

For example, in the real world, we might ask what is the probability that a factory worker suffers a work-related injury in the
past year? This question combines being a factory worker—a machinist, a textile worker, a structural iron and steel worker, or a
mechanic (among many others)—and being a person who has suffered a work-related injury—a mailman slipping as he
delivers mail, an assembler suffering repetitive stress injury, a construction worker falling from scaffolding (and many other
injuries).

The Venn diagram can help convey the different categories. It shows a visual representation of all the possible results. The
area labeled A
represents A
occurring—such as someone being a factory worker in the past year—while the area labeled B
represents B
happening—such being a person who suffered from a workplace injury in the past year. The darker area represents both
A and
B occurring, that is the set of all factory workers who suffered work-related injuries last year.

Copyright © 2019 MindEdge Inc. All rights reserved. Duplication prohibited.

Click on the titles below to see the categories with the Venn diagram.

Universe

The universe is the entire category being considered. It is not a given, but defined, just as the categoriesA
and B
are defined.

In our example of factory workers in the past year A(

) and people who have suffered from workplace injuries in the past year B (
), the universe would be "people who were employed in the past year." The universe is defined based on the question
that researchers are interested in studying. If we were calculating probabilities such as the probability of factory
workers being injured at work compared to the probability of other workers being injured at work, it would make sense
for the universe to be "all employed people within the past year." In the diagram below, the orange shaded area
represents all the elements in the universe.

Copyright © 2019 MindEdge Inc. All rights reserved. Duplication prohibited.

 Of course with "all people employed within the past year," as the universe, we could define other groups, such as
"employed people who suffered injuries at home" or "employed people who made less than $20, 000
a year." These might be important categories for other types of research questions.

"Not"

The word "not" before a category is everything outside that category.

In the Venn diagram above, the white space aroundA

and the parts of B
that are not in A
represent "not A
," that is the outcomes that are not inA
, but are still in the universe. "NotA
" in our real-world example is "Americans employed last year as something other than factory workers." Teacher,
postal worker, librarian, business woman, etc. all belong to "not A
." Note that it includes people in category B
who are not in A
, such as librarians who were injured by falling bookshelves. However, it is best not to explicitly considerB
when trying to decide if something is not A
. It's enough to ask "is this person someone who was employed last year in a type of factory work?". If not, she
belongs to "not A
." In the diagram below, the orange shaded area represents all the elements in "notA
."

Copyright © 2019 MindEdge Inc. All rights reserved. Duplication prohibited.

Not B
refers to outcomes that are not in the category of B but still in the universe.

In our real-world example, not B

would be employed people who were NOT injured at work last year, including all factory workers who were not injured
at work, as well as all doctors, nurses, garbage workers, and truck drivers who were not injured at work. In the
diagram below, the orange shaded area represents all the elements in "not B
"

Copyright © 2019 MindEdge Inc. All rights reserved. Duplication prohibited.

A
and B
(intersection)

It is easy to get confused with the expression "A

and B
." The word "and" in English usually indicates a more expansive possibility. You might say "don't bring just Jack to the
party; bring Jack AND Jill!" In probability, on the other hand, the word "and" indicates greater restriction. We are
considering only those elements in both A
and B
. This is why the keywords for this category are "and" and "both." The intersection ofA
and B
will be the same size or smaller—not larger—than the sets of A
or B
individually.

In our real-world example, we are considering only factory workers who suffered workplace related injuries. We are
excluding factory workers who did not have any injuries and librarians who did. A
and B
includes fewer elements than either A
or B
alone. A
and B
is called the intersection of the two categories. It is represented in pale purple below.

Copyright © 2019 MindEdge Inc. All rights reserved. Duplication prohibited.

A
or B
(union)

"A
or B
" is what in English we would normally define as "A
and B
," that is everything that is in A
and everything that is in B
. You can remember the probability meaning of "or," the union, by realizing that the "or" means an outcome is in the
category as long as it is in either A
or B
. The key words for this category are "or" and "either." A
or B
is called the union of A
and B
. The total orange area indicates the union.

Copyright © 2019 MindEdge Inc. All rights reserved. Duplication prohibited.

7.07 Complementary Events
Complementary Events

Sum of Probabilities

Remember that the probabilities of each individual outcome in the sample space sum to 1. For example, suppose the
experiment is "being born on a day of the week." The sample space would be Monday, Tuesday, Wednesday, Thursday,
1
Friday, Saturday, and Sunday. The probability of being born on each of those days is 7
. Since there are 7
days, the sum of all the probabilities is

1 1 1 1 1 1 1
7 + 7 + 7 + 7 + 7 + 7 +7
7
= 7
=1

1
Similarly, if the experiment is flipping a coin, the events are "heads" or "tails." The probability of flipping a heads is2
1
. The probability of flipping a tails is 2

Copyright © 2019 MindEdge Inc. All rights reserved. Duplication prohibited.

. The sum of these two probabilities is

1 1
2 +2 =

2
2 =

Complements

The opposite of an event happening (i.e. the event not happening) is called the complement of the event. The sum of the
probability of an event and the probability of its complement is always equal to 1
. For example, the complement of flipping heads is flipping tails. As we showed above the sum of flipping heads and flipping
tails is 1
.

Here are some examples of complementary events.

Complementary events are those that do not have any common outcomes, and the union is the whole
universe. One outcome is the event itself, and the other outcome is the event not happening. For example, flipping a
coin and it landing on either heads or tails — those are the only two possible outcomes, so the two events are
complementary.

If there are outcomes in the universe that are not in either event, then the two events are not complementary.
For example, picking either a large shirt or a medium shirt are not complementary events. There are shirts in many
sizes! You could pick a large, medium, small, extra large, etc. So picking a large shirt or a medium shirt are not

Copyright © 2019 MindEdge Inc. All rights reserved. Duplication prohibited.

 complementary.

The events must be properly defined in order to have complementary events. For example, "rolling a die and
getting an even number" would not be a complete instance of complementary events (as this is only one event).
However, "rolling a die and getting an even number" and "rolling the die and getting an odd number" are
complementary events; the events are clearly defined.

Example: Probability of Complementary Events

Let Let
A = the day is a weekday B = the day is a weekend
What is What is
P(A)? P(B)?
The formula for probability is
Number of desired events
Total number of events
The number of weekdays is 5 The number of weekend days is 2

The total number of days in the week is 7 The total number of days in the week is 7

() ()
5 2
P A = 7 P B = 7

5 2
7 +7 =1

Notice from the example above that since the sum of the probability of an event and its complement is1
, the probability of 1
minus the probability of that event is the probability of the complement.

Another way to write this is:

P(A) + P(not A) = 11 - P(A) = P(not A)1 - P(not A) = P(A)

Example
Returning to Bob's suits:

If Bob likes the suits equally, what is the probability he will choose to wear blue?

(
(3 blue suits ) 3
P Bob wearing a blue suit)= (5 suits total) = 5

What is the probability of Bob not wearing a blue suit?

The probability of Bob not wearing a blue suit is equal to

1 - P(Bob wearing a blue suit)
. Since P(Bob wearing a blue suit)
3
is 5
, the probability of Bob not wearing a blue suit can also be calculated as,

Copyright © 2019 MindEdge Inc. All rights reserved. Duplication prohibited.

 3 5 3 2
1- 5 = 5 - 5 = 5

( )
2
P Bob not wearing a blue suit = 5

The fact that the probabilities of an event and its complement sum to 1 is not just interesting; it can be useful with some real-
world problems. In some cases it may be difficult to calculate the probability of an event, but easier to calculate the probability
of its complement. In these cases 1 - P(complement)
might provide an easy calculation to a difficult problem.

Complement Rule Part Two

One example of how the complement rule can be helpful is studying problems with the phrase "at least one." For example,
remember our previous example of the family planning to have three children? There were eight possible outcomes in terms of
gender. What is the probability that the family will have "at least one" girl? The complement of "at least one X
" is "no X
." The complement of having "at least one" girl is having "no girls." It turns out that calculating the probability of having "no
girls" is very easy: it's the same as the probability of having three boys.

Using the complement rule

() ( ) ( ) ( )
1 8 1 7
1 - P A = P not A 1 - P having no girls = P having at least one girl 1 - 8 = 8 - 8 = 8

7
The probability that a family has at least one girl is 8
.

More generally, because "at least one X

" and "no X
" are complements, it is always true that

P(at least one X) = 1 - P(no X)

7.08 Disjoint Events

Disjoint Events
Before we introduce formulas for probabilities, we will continue to define different categories of events. Disjoint events have
special characteristics that make their intersection and union unlike those of other events.

If two events cannot both occur at the same time, they are called disjoint. This concept applies to any number of multiple
events. For example, a coin cannot be both "heads" and "tails" at the same time. The three statements below are disjoint:

Examples of Disjoint Statements

I was born in January.

I was born in March.
I was born in May.

No two of those statements can both be true.

A population and a sample can be disjoint as well, but they shouldn't be. For example, suppose you are studying the

Copyright © 2019 MindEdge Inc. All rights reserved. Duplication prohibited.

prevalence of obesity in pre-school children. You would not survey the local high school in order to gather data about the
population of pre-schoolers. That would make your population and your sample disjoint, which is nonsensical if you're trying to
use your sample to learn something about your population. On the other hand, taking two disjoint samples can make sense: in
your study of obesity in pre-schoolers, you might take disjoint samples—one in California, one in Massachusetts.

The following Venn diagram shows how disjoint events differ from other events: there is no intersection, no common elements
between the two. For example, there is no intersection among "people born in January" and "people born in March."

Disjoint A and B (intersection)

It is relatively easy to calculate the probability of two or more disjoint events. If there is no overlap, there is no
possibility of belonging to both sets. The probability that someone is born in both March and January, for example, is
0
.

This is true of the intersection of all disjoint events:

If A and B are disjoint,

P(A and B) = 0
Disjoint A or B (union)

Calculating the probability of the union ofA

and B
is also logical. Since there is no overlap, the probability of being in either one or the other is the sum of their
individual probabilities.

What is the probability that someone is born in January or March, for example? The sample set of "month of birth" is
comprised of the 12
months of the year.

If you think about it, there are 2

ways to be born in January or March. You can be born in January (1
chance out of 12
months) or you can be born in March (1
chance out of 12
months). These 2
2
chances lead to a probability of 12
.

Copyright © 2019 MindEdge Inc. All rights reserved. Duplication prohibited.

 ( ) ( )
1 1
P being born in January = 12 P being born in March = 12

The probability of being born in January OR March is

1 1 2 1
12 + 12 = 12 = 6

This is called the addition rule of probability:

If A and B are disjoint,

P(A or B) = P(A) + P(B)

Let's look at another example.

Slide 1

Drawing from the example of three children, suppose you want to calculate the probability of having a boy-
girl-boy (which we'll call event A
) or having a girl-boy-girl (event B
).

Slide 2

These events are clearly disjoint. There is no way to do both. To calculate the probability of one or the other,
you use the addition rule.

Copyright © 2019 MindEdge Inc. All rights reserved. Duplication prohibited.

 Slide 3

Disjoint and Complementary

Note that events can be disjoint without being complementary. An example of events that are both disjoint and complementary
are "flipping heads" and "flipping tails." Together these events constitute the universe of possible events, so they are
complementary. You also cannot have them at the same time, which means they're disjoint. But the events "being born on a
Wednesday" and "being born on a Friday" are disjoint without being complementary. There are other possibilities so even if
you are not born on a Wednesday, that does not necessarily mean you are born on a Friday.

Important Numerical Phrases

When understanding an event, some phrases can be challenging to interpret until you get used to them. For example, when
the Height of Buildings Act was originally passed in 1899, it limited the heights of any type of building in Washington, D.C. to
no higher than 110
feet. For any given building, a height of 110
feet was the limit. One category for measurement, therefore, might be anything "above 110
feet," as those heights are not permitted. A commercial building with a height of 109
feet would not fall into the category of "above 110
feet," while a commercial building that's 111
feet tall would. So the event "being a commercial building that's only 109
feet tall" and "being in violation of the Height of Buildings Act of 1899 (that is having a height above 110
feet) are disjoint.

Similarly, a category might have a range of values or require one of a set of conditions. In all these cases, pay attention to
whether the value you are considering belongs to the category or is disjoint.

Interactive Drag and Drop Game: Is The Event Disjoint or Not?

Disjoint Probability

Notice that when asking about disjoint events, we naturally employ the word "or." Questions that ask about the probability of
one or more disjoint events, always use an "or" and can be calculated using the formula:

P(A or B) = P(A) + P(B)

Exercises
Using the addition rule above, calculate the probabilities of the following disjoint events. (Assume all the events in the sample
space are equally likely.) Enter the result as a fraction. Use the / sign to indicate division. Always reduce fractions to lowest

Copyright © 2019 MindEdge Inc. All rights reserved. Duplication prohibited.

terms.
1. What is the probability that a die shows a2
or a 3
when rolled?
2. What is the probability that a person was born on a Tuesday, Thursday, or a Friday?
3. What is the probability that a coin shows either a heads or a tails?
4. What is the probability that a traffic light is green or yellow (assuming each light lasts for the same duration)?
5. If balloons come in equal numbers of red, green, yellow, orange, purple, pink, gray, and peach what is the probability that
someone gets a ballon that is NOT red?
6. What is the probability that a 10
-sided fair die shows a multiple of3
or 10
?
7. You roll two six-sided dice. What is the probability of rolling a sum less than5
or greater than 10
?
8. What is the probability of winning a raffle with1, 000
tickets if one family member buys 3
raffle tickets and another family member buys 2
tickets?
9. Out of class 20
students, what is the probability that you or your 3
friends are picked first?
10. What is the probability of selecting a weekend day or Memorial Day (Monday) in May if the first day is a Sunday?

7.09 Independent Events and The Multiplication Principle

Independent Events and The Multiplication Principle

Is there a way to determine the probability of a complex series of options, such as the having-three-children situation
discussed on the earlier page, without taking the time to draw out a tree or make up a table? The answer is yes, but before you
learn the process, you need to know another definition.

Independent events are those that are not affected by other trials or events. For example, if you were to flip a coin once, that
first result (either heads or tails) would not have any impact when the coin is flipped a second time — the first event gives no
indication of what could result from the second event. A more real-world example is two people who don't know each other
having heart attacks. If someone has a heart attack, that has no effect on the probability the second person will have a heart
attack.

When considering whether two events are independent or dependent, think of one event in isolation, then think of that event
again in relation to the second. Let's do this with two randomly selected people and the likelihood that they will graduate from
college. First, consider the first person in isolation. According to the most recent U.S. Census data, 68 %
of the population graduates from college. Now, suppose the second person also graduates from college, does this change the
probability for the first? No! Since the second person has no effect on the first, these events are independent.

The events we have discussed in this module have been largely independent. Here is a quick example of two events that are
not independent.

Example: An Event That is Not Independent

In a candy container, there are five white chocolates and ten dark
chocolates. If you pick a chocolate at random out of the jar, the
probability of what you will get on the next pick is affected.

If you pick a dark chocolate on your first pick, there will be fewer dark
chocolates the next time you pick, lowering the probability you get a
dark one on the second draw. On the other hand, if you pick a white
chocolate on your first pick, there will be fewer white chocolates the
next time you pick, lowering the probability you will get a white on the
second draw.

Copyright © 2019 MindEdge Inc. All rights reserved. Duplication prohibited.

Unlike the coin toss, what you get in the first trial is not independent
from what you get in the second trial. Another, more real world example, is the event "smoking cigarettes" and "getting lung
cancer." The occurrence of the first event affects the probability of the second event.

Disjoint Events: Dependent

Disjoint events are dependent. For example, if you know someone is born on a Wednesday, you also know that he or she
1
cannot be born on a Friday. If you know someone is not born on a Wednesday, there is a greater possibility, 6
1
rather than 7
, that he or she is born on a Friday. Because the occurrence of one affects the probability of the other, these events are
dependent.

Independent Events: P(A and B)

You may have noticed a pattern when we were constructing the tree for having three children. For each new option we
had to add an additional set to each of the previous options. This, in essence, is multiplication. At each point in the
tree the probability of one particular outcome, say "having a boy," is equal to 1
divided by the total number of possibilities, 2
. As the number of options increase, the new probability is captured by multiplying the previous probability by the
probability of the new independent event in the sequence. This is the multiplication rule for independent events.

The Multiplication Rule for Independent Events

When two events are independent,

P(A and B) = P(A) × P(B)

Example
1
For example, the probability of rolling a one with a six-sided die is 6
1 1 1
. When rolling two dice with six sides each there are 6 × 6 = 36
probability of rolling double ones. You can verify this probability on the table below. Notice there is only one cell with
double ones and there are 36
(6 × 6
1
) cells in total. Therefore, the probability of double ones is 36
.

Copyright © 2019 MindEdge Inc. All rights reserved. Duplication prohibited.

 1 1 1 1
Likewise, the probability of rolling triple ones when rolling 3 dice is 6 × 6 × 6 = 216
.

Example 2
Suppose there are two different kinds of protection gloves, A and B used by assembly line workers. A costs less but
has a defective rate of 4. 2 %
. The more expensive brand, B, costs more, but has a defective rate of1. 7 %
. If 3 %
of the actions performed by the assembly line workers involve potentially hazardous materials, what is the percentage
chance of exposure for each brand of glove?

Using the multiplication rule for independent events, you multiply 4. 2 % × 3 %

to find the probability of exposure for brand A. You perform a similar calculation to find the probability of exposure for
brand B.

Probability of exposure for brand A: 0. 042 × 0. 03 = 0. 00126

Remember to convert probabilities to decimals before multiplying and remember to count the total number of places
after the decimal point to obtain the correct answer.

Probability of exposure for brand B: 0. 017 × 0. 03 = 0. 00051

You can convert both decimal probabilities back to percentages if you feel more comfortable interpreting data in that
form. Brand A has a 0. 00126
probability of exposing an assembly line worker to a hazardous material, which is 0. 126 %
. Brand B has a 0. 00051
probability of exposing an assembly line worker to a hazardous material, which is 0. 051 %
.

Copyright © 2019 MindEdge Inc. All rights reserved. Duplication prohibited.

Independent Events: P(A or B)

The probability of either of the two events occurring is given by a general formula. Consider again the union of two
events, A or B.

The red line around the space of A, B and their intersection indicates the space of events characterized by "or."

A co-worker jokes that Bob always wears black. Referring again to the table, the probability of Bob wearing a black
suit OR a black shoes is
11
15 . There are 11 cells that have either a black suit or black shoes (or both) and
11
15 cells total. 15
is the probability of either event happening, including the situation in which both events happen.

Bob's Outfits
black shoes black shoes brown
1 2 shoes
black suit, black suit,
black suit,
black suit black shoes black shoes
brown shoes
1 2
brown brown suit, brown suit, brown suit,
suit black shoes black shoes brown shoes
1 2
blue suit 1 blue suit 1
blue suit 1
, ,
blue suit 1 ,
black shoes black shoes
brown shoes
1 2
blue suit 2 blue suit 2
blue suit 2
, ,
blue suit 2 ,
black shoes black shoes
brown shoes
1 2
blue suit 3 blue suit 3
blue suit 3
, ,
blue suit 3 ,
black shoes black shoes
brown shoes
1 2

How can we get this answer without creating a whole table?

Copyright © 2019 MindEdge Inc. All rights reserved. Duplication prohibited.

 To calculate an "or" problem, it makes sense to add the probability of both events. The probability of one of two (or
more) events occurring is just the probability of each event added together. But notice that when you add all the times
Bob wears black shoes and all the times Bob wears a black suit, you are counting the cells where both occur twice.

( ) ( ) ( )
1 2 1 2 3 10 13
P Bob wears a black suit = 5 P Bob wears black shoes = 3 P Bob wears a black suit or black shoes = 5 + 3 = 15 + 15 = 15

Notice that this number overstates the union by two cells.

Where did we go wrong? We counted the probability of wearing both a black suit and black shoes twice. You can see
that in the table above two cells have two orange terms.

General Addition Rule

P(A or B) = P(A) + P(B) - P(A and B)

By subtracting the intersection, P(Bob wears a black suit and Bob wears black shoes)
, we avoid counting that area twice in the sum ofP(Bob wears a black suit)
and P(Bob wears black shoes)
. So the formula is to add P(Bob wears a black suit)
and P(Bob wears black shoes)
, and then subtract P(Bob wears a black suit and black shoes)
.

Another way to understand it is to look at the Venn diagram above. To find the probability of both circles, you add the
"area" of each circle, but then, notice you've counted the area of the intersection twice (once for the blue circle and
once for the yellow circle), so it will need to be subtracted to compensate. For disjoint events, there is no area of
intersection, so no need to subtract it.

The formula above is the formula for events that are not disjoint. Remember that for events that are disjoint, the
general rule simplifies to P(A or B) = P(A) + P(B)
since there is no area of intersection.

Worked Problems
Answer in
Question Formula
%
P(A or B) = P(A) + P(B) - P(A and B)

If the probability of a sunny day is 0. 7 Since "calendar date being an odd number" and 85 %
and the probability of the date on the calendar "being a sunny day" are independent events, chance of it being sunny
being an odd number is 0. 5 P(A and B) = 0. 7 × 0. 5 = 0. 35 or being an odd
, what is the probability of the calendar date being . numbered day on the
odd or it being a sunny day? calendar.
Calendar date being an odd number or being a
sunny day = 0. 7 + 0. 5 - 0. 35 = 0. 85

Kelly gets a donut on the way to work when she

has time and the donut shop has chocolate
donuts. If the probability of her having time by We are looking for an intersection here, an "and" 37. 5 %
getting the earlier train is 75 % problem. chance of getting a
and the probability of the shop still having
P(A and B) = P(A) × P(B) = 0. 75 × 0. 5 donut.
chocolate donuts is 50 %
, what is the probability that she gets her
chocolate donut?

In a certain game, you roll a fair, six-sided die These are disjoint events.
33 %
Copyright © 2019 MindEdge Inc. All rights reserved. Duplication prohibited.
In a certain game, you roll a fair, six-sided die These are disjoint events.
once. You win if you roll a 3 33 %
chance of winning

( )
or a 4 1 1
. What is the probability of winning this game? P A or B = 6 + 6

P(A or B) = P(A) + P(B) - P(A and B)

.

There are 2
numbers less than 3
, so the probability of event A
2
is 6
1
or 3
.

In another game, you win if you roll a number less There are 3
than 3 67 %
()
1
or if you roll an even number. What is your numbers that are even, so P B = 2 chance of winning
probability of winning this game?
.

But the number 2

is both less than 3

( )
1
AND even, so P A and B = 6

The formula gives us

1 1 1 2
3 + 2 - 6 = 3

This is an "and" question, the probability that he

works from home (event A) and it is sunny outside
(event B).
Ricky randomly works 3
out of every 5
()
3
24 %
days from home. In the area where he lives, it is P A = 5 = 0. 60
of the time, Ricky works
sunny outside 2
from home and it is
out of every 5
sunny outside.

()
days (on average). What is the probability that he 2
works from home and it is sunny outside? P B = 5 = 0. 40

P(A and B) = 0. 60 × 0. 40 = 0. 24

This is an "or" question, the probability that he

works from home (event A) or it is sunny outside
(event B).
Ricky randomly works 3

()
3
out of every 5 P A = 5 = 0. 60 76 %
days from home. In the area where he lives, it is
of the time, Ricky works
sunny outside 2
from home or it is sunny
out of every 5

()
2
outside.
days (on average). What is the probability that he P B = 5 = 0. 40
works from home or it is sunny outside?

P(A or B) = P(A) + P(B)– P(A and B) =

0. 60 + 0. 40– 0. 24 = 0. 76

Copyright © 2019 MindEdge Inc. All rights reserved. Duplication prohibited.

7.10 Frequency Tables
Frequency Tables

Career Connections

Frequency Tables

A frequency table is a way to organize raw data so it can be easily analyzed. Numbers of people
(or objects) fall into different categories, allowing you to compare the effect of the categories.

Managers may use frequency tables when evaluating processes. For example, suppose you are
running a manufacturing line, and you want to discover the most accurate method of quality
control. You might set up a test run with visual controls (looking at the object for defects), manual
controls (examining the object for defects), and automated controls (a machine measures the
object for defects). You then measure the number of defects that were detected and the number that were undetected by
each of the control processes. You can then analyze your numerical results.

The advantage of a frequency table is you have the raw data in front of you in a form that is easily understood. You can also
add data. Suppose you wonder if a combination of automated and visual controls might catch more defects. You can add
another row to the frequency table to analyze this new process.

A frequency table (two-way table) is a table consisting of frequency counts of two categorical variables. One variable is called
the row variable, and the other variable is called the column variable. These tables are very helpful when it comes to
analyzing data.

In the examples below, the entries in the table are the counts of cases in the sample with certain results for each of the two
variables. In the following two-way table, the two categorical variables are Section (the row variable) and Student's Status (the
column variable).

Student Status:
Passed Failed
Section Section A 375 25
Section B 550 50

Table Construction

In two-way tables, the cells represent the number of objects (usually people) who fall into the two categories signaled by the
cell's row and column. To read a table, match each cell with its row and column. So in the table above, there were 375
people who are in Section A and passed. In the next cell in the same row, there were 25
people who are in Section A and who failed.

An important aspect of two-way tables is that each person can fall into only one cell. In the example above, someone cannot
be in both Section A and Section B. Obviously, someone cannot both pass and fail. To find the total number of people involved
in the study, you can add up the numbers in each cell. You will not double count anyone, because everyone appears only once
in the tally.

Two-way Table Analysis

There are many ways to analyze the data in a two-way table. To assist in the analysis, we should add a row and a column to
record the totals for each:

Student Status:
Passed Failed Total

Copyright © 2019 MindEdge Inc. All rights reserved. Duplication prohibited.

 Section Section A 375 25 400
Section B 550 50 600
Totals 925 75 1000

With the totals we can calculate a number of interesting results that will help us get a better handle on the situation.

Sample Relative Frequency Questions and Explanations

Question Explanation Probability
400
What percent of the students were in Section A? 1000
40 %
= 0. 4
600
What percent of the students were in Section B? 1000
60 %
= 0. 6
925
What percent of the students passed? 1000
92. 5 %
= 0. 925
75
What percent of the students failed? 1000
7. 5 %
= 0. 075
What percent of the students in Section A passed?
To answer this question, notice that there are 375
400
93. 75 %
400 students in Section A, and = 0. 9375
375 of them passed.
What percent of the students in Section B passed?
To answer this question, notice that 550
600
91. 67 %
550 students passed out of a total of = 0. 9167
600 students.
375
Of the students who passed, what percent were in Section A? 40. 54 %
Section A = 925
Section A: Section A ÷
Total number of students who passed

Section B?
Section B: Section B ÷ 550
Total number of students who passed 59. 46 %
Section B = 925

Frequency Tables and Independent Events

Remember the definition of independent events, that the occurrence of one does not affect the probability of another? In
practice many times researchers use the relative frequency of events to determine their independence. A researcher might
wonder if gender affects the probability of receiving a speeding ticket. They use study results to determine independence. If
men and women receive a speeding ticket equally frequently, we would determine that gender is independent of receiving a
speeding ticket.

Employment and Drinking Tea

Consider the following table showing tea consumption and whether an employee was laid off at a major corporation, during a
period of lay-offs. We might use the table to ask if tea consumption and being laid off are linked or independent.

Status:
Not Laid Off Laid Off Total
Tea Consumption None or 1 514 285 799
Cup per Day
2 129 72 201
or More Cups per Day
Totals 643 357 1000

How can we use the table to determine whether drinking tea and being laid off are dependent or independent events? If
drinking tea and being laid off are independent events, the occurrence of one will not affect the probability of the other.

Copyright © 2019 MindEdge Inc. All rights reserved. Duplication prohibited.

How do we estimate the probability of "drinking2
or more cups of tea a day"? Find the number of heavy tea drinkers and divide it by the total number of people. This will give us
201
the relative frequency of being a heavy tea drinker: 1000 = 0. 201
or 20. 1 %
.

What is the probability of drinking 2

or more cups of tea if a subject is laid off? If the events are independent, being laid off should not affect the probability of
drinking tea. Here we look at the numbers in the "laid off" column: 285
and 72
. They total to 357
. Of the 357
laid off employees, 72
drink 2
72
or more cups of tea a day. The relative frequency of drinking a lot of tea while being laid off is 357 ≈ 0. 20167
(rounded to the nearest hundred thousandth), or 20. 2 %
. The relative frequencies are very similar, suggesting that the probabilities are the same and these two events, therefore, are
independent.

7.11 Conditional Probability

Conditional Probability

Dependent Events

Just as events can occur independently, there are also those whose occurrence can affect the probability of one another. If
there is one winning ticket in a bowl and 49 losing tickets, each time someone chooses (and then discards) a losing ticket,
there are fewer losing tickets. Therefore, the probability of picking the winning ticket increases slightly with each losing ticket
picked out of the bowl. Events in which the probability changes depending on other events are categorized as dependent
events. Conditional probability is used to calculate these kinds of events. Conditional probability is the probability of an event
occurring (such as picking the winning ticket), given that another event has already occurred (such as someone else having
already picked one of the losing tickets).

Examples of Dependent Events

Being dealt an ace and drawing a second ace

Opening an attachment to an email from an unknown sender and getting a computer virus

Breaking the law and getting arrested

Buying a lottery ticket and having a chance to win the lottery

Being read to as a toddler and knowing more than800

words by age 3

Spending more than 2

hrs each day in a car and having been in an car accident during the last year

Conditional Probability

Conditional probabilities represent a prevalent category of probabilities. From card games to medical conditions, people want
to know how one event or condition affects the probability of another event or condition. There is a special notation to indicate
conditional probability, the probability that a second event B
will occur, given the occurrence of event A
. We write P(B | A)
, which means "the probability of event B

Copyright © 2019 MindEdge Inc. All rights reserved. Duplication prohibited.

, given event A
has occurred." The probability of event B
occurring is dependent on whether event A
has occurred. Please note,

P(B | A)
is read "The probability of
B, given
A". For example, if B
is "a person is a CEO" and A
is "that person is an MBA," P(B | A)
is the probability of "a person is a CEO" given "that person is an MBA." B
is the event we are interested in; A
is the event that might affect it.
P(B | A)
indicates conditional probability. If A
and B
are independent events, P(B | A) = P(B)
.
The |
between B
and A
is a vertical bar indicating a conditional probability. It is a symbol not a division sign.

Example
A cookie jar has four sugar cookies and six chocolate chip cookies. With conditional probability, we're looking to find the
probability of choosing certain cookies in different circumstances.
4
What is the probability that a sugar cookie is chosen first? There's 10
2
or 5
chance a sugar cookie is chosen first, but what is then the probability
of choosing another sugar cookie?

There is one less sugar cookie in a jar that now has only9
3
cookies. Picking a second sugar cookie has a probability of 9
1
or 3
. Since it is affected by the event before, the probability of picking
another sugar cookie is smaller. The probability of this event is
dependent on the event prior. The probability of picking a sugar
1
cookie on the second draw, given you have already picked a sugar cookie on the first draw is 3
.

If the first cookie chosen is chocolate chip, the cookie jar would then have all four sugar cookies remaining, among the nine
cookies left in the jar. Therefore, probability of choosing a sugar cookie on the second draw, given you have already picked a
4
chocolate cookie is 9
.

Notice how the numbers change and will continue to change as the events occur — because the circumstances are changing
and the events are dependent upon one another, the probabilities also change.

Example 2
Suppose there are two convenient parking spaces at a company building and the remaining eight are far from the door. There
are two elderly employees at the company and eight middle-aged employees. The arrival time of employees in the morning is
random.

Copyright © 2019 MindEdge Inc. All rights reserved. Duplication prohibited.

 What is the probability that an elderly employee gets the What is the probability that an elderly employee gets the
second space, given that a elderly employee has already second space, given that a middle-age employee has
gotten the first space? gotten the first space?
Let A =
"An elderly employee gets the first near parking spot"
Let B =
"An elderly employee gets the second near parking spot"
P(B | A) = P(B | not A) =
Because one employee has parked, there are now nine Because one employee has parked, there are now nine
employees looking for spaces. Because one of the elderly employees looking for spaces. Because neither of the elderly
employees has already gotten the first spot, there is one left. employees has gotten the first spot, there are two left
1 2
9 9

Notice in the example above, the probability ofB

changed if A had occurred. P(B) ≠ P(B | A)
shows that A
and B
are dependent events.

Career Connections

How Circumstances can Affect Student Learning

Here's something important to teachers: It is known that certain circumstances affect student
learning. For example, if about 85 %
of a class can master a certain concept in one day, you might expect that students with a learning
disability have a lower chance of learning the same skill in one day. This is an example of
conditional probability. Said another way, a probability can vary depending on conditions or
circumstances.

How can this help you in the classroom, though? First, it gives you a way to quantify some of things that classroom experience
teaches you (such as that, in general, students with learning disabilities need more time for concepts). Does that mean you'll
be figuring out these probabilities in your classroom, though? Probably not. But, thinking of learning in this way can give you a
different angle to improve learning for all your students in all kinds of different circumstances.

It can also help you think of educational research from other perspectives, thus helping you think critically about applying
educational research findings to your classroom. Sometimes it's not about how we use a statistical concept directly;
sometimes it's more about how the concept helps us see the world differently.

Interactive Drag and Drop Game: Are the Two Events Dependent or Independent?

7.11.1 Probability of Dependent Events

Probability of Dependent Events
Thus far we have calculated the probability of a second event occurring, given a first event. Often we want to determine the
probability of both events occurring or some other situation. Here we give some general rules for calculating conditional
probabilities

Conditional Probability Rule

Conditional Probability Rule

Copyright © 2019 MindEdge Inc. All rights reserved. Duplication prohibited.

 (|)
P(A and B )

P B A = P(A)

Let us illustrate this rule by using a very simple example. Imagine you are flipping a coin twice. LetA =
"getting a head on the first flip" and let B =
"getting two heads." Let us look at a number of categories related to probability:

1. The sample space of this event is HH, HT, TH, TT

2. P(A)
is 2
possibilities out of 4
1
or 2

3. P(B)
is 1
possibility out of 4
1
or 4

4. A
and B
, the intersection, is the set of elements that belong to both categories, which is just the element HH. So the

( )
1
P A and B = 4

(|)
1
5. If you have already flipped a head, you have two possibilities for the second flip, heads and tails.P B A = 2

So substituting the probabilities above into our general conditional probability formula

(|)
P(A and B)

P B A = P(A)

gives
1
4
1
1
2 = 2

This very useful formula generally cannot be verified by writing out the entire sample space. Instead, we often use it
with probabilities that we are given.

Example
A small business is trying to learn more about their customers. Suppose8. 6 %
of their customers are homeowners who heard of the business from a TV advertisement. 18 %
of all their customers are homeowners. What is the probability that a home-owning customer heard of the business
from a TV advertisement?

Let A =
a home-owning customer; Let B =
hearing of the business from a TV advertisement

( ) () (|) ( )
0 . 086

P A and B = 0. 086P A = 0. 18P B A = 0 .18 P hearing from a TV ad given homeowner = 47. 8 %

Interpretation: close to 50 %

Copyright © 2019 MindEdge Inc. All rights reserved. Duplication prohibited.

 of the home-owning customers heard of the business from a TV advertisement.

Dependent Events: P(A and B)

The conditional probability rule gives us the formula we can reformulate to calculate the probability that both events
occur.

Remember that

(|)
P(A and B)

P B A = P(A)

We can multiply both sides by P(A)

to solve for P(A and B)
:

The General Multiplication Rule

P(A and B) = P(A) × P(B | A)

Returning to our earlier example of the cookies, remember that the cookie jar has4
sugar cookies and 6
chocolate chip cookies.

Let A =
"choosing a sugar cookie first." Let B =

()
4
"choosing a second sugar cookie." P A = 10

2
or 5

(|)
3
. Remember that once you've chosen a sugar cookie, there is one fewer and one fewer of all the cookies.P B A = 9

1
or 3
.

Therefore:

( )
2 1 2

P A and B = 5 × 3 = 15

2
The probability of choosing two sugar cookies from the bowl is 15
.

Example
Veryfake University is examining their enrollment practices. Of accepted students with a
3.0
or higher High School GPA, 26 %
end up enrolling. Of enrolled students, 68 %
of accepted students graduate in four years. What is a student's probability of having a 3.0
or higher High School GPA and graduating in four years?

Let A =
being an accepted student with a 3.0

Copyright © 2019 MindEdge Inc. All rights reserved. Duplication prohibited.

 or higher High School GPA
Let B =
graduating in four years

P(A) = 26 %

P(B | A) = 68 %

What is the P(A and B)

?

P(A and B) = P(A) × P(B | A)P(A and B) = 0. 26 × 0. 68 = 0. 176

There is a 17. 68 %
probability of having a 3.0
or higher High School GPA and graduating in four years.

Example
In Fictionaltown, the chances of a home being flooded in any given year are1
in 88
. The probability of having one's home flooded if it already had a flood last year is7 %
. Given these statistics, what is the probability that a random home was flooded both last year and this
year?

Let A =
being flooded last year
Let B =
having a flood for the second year in a row, this year

() (|)
1

P A = 88 = 0. 01136P B A = 7 %

What is the P(A and B)

?

P(A and B) = P(A) × P(B | A)P(A and B) = 0. 01136 × 0. 07 = 0. 000795

There is a 0. 08 %
chance that a random home was flooded both last year and this year.

Dependent Events: P(A or B)

Earlier we learned how to calculate the probability of the union of disjoint events:

P(A or B) = P(A) + P(B)

We also learned a "General Addition Rule" for independent events. The rule for dependent events is the same, and is
reproduced here:

General Addition Rule

P(A or B) =
P(A) + P(B) - P(A and B)

Let us return to our simple example to see how the general addition rule works for dependent events as well. Recall
that we are flipping a coin twice and A =
"getting a head on the first flip" and B =
"getting two heads."

Copyright © 2019 MindEdge Inc. All rights reserved. Duplication prohibited.

 P(A or B)
is the probability that either a head is flipped first or two heads are flipped. Clearly, event B
includes event A
; if we get two heads, we must have flipped heads on the first flip. There are two cases where we flip heads on the first

( )
1
flip: HH and HT. Therefore P A or B = 2

Notice how this probability follows the general addition rule:

( ) () () ( )( )
1 1 1 1
P A or B = P A + P B - P A and B P A or B = 2 + 4 - 4 = 2

Example
Of accepted students with a 3.0
or higher High School GPA, 26 %
end up enrolling. Of enrolled students, 17. 68 %
had a 3.0
or higher High School GPA and graduate in four years. Of all Veryfake students, 25 %
graduate in four years. What is the probability, among Veryfake U students, of having a
3.0
or higher High School GPA or graduating in four years?

Let A =
having a 3.0
or higher High School GPA
Let B =
graduating in four years

P(A) = 26 % P(B) = 25 % P(A and B) = 17. 68 %

What is the P(A or B)

?

P(A or B) = P(A) + P(B) - P(A and B)P(A or B) = 0. 26 + 0. 25 - . 1768 = 0. 3332

There is a 33. 32 %
chance, among Veryfake U students, of having a 3.0
or higher High School GPA or graduating in four years.

We have seen some examples of how to use the formulas above. Here is a process to use to solve conditional probability
problems:

To find probabilities:

1. Create events A
and B
based on the givens of the problem
2. Establish what kind of probability you will be finding, such as P(A and B)
or P(A | B)
, depending on what the problem is asking
3. If necessary, determine if A
and B
are dependent or independent events

Copyright © 2019 MindEdge Inc. All rights reserved. Duplication prohibited.

 4. Write out the appropriate formula
5. Determine the probabilities needed for the formula (or take them from the givens of the problem)
6. Calculate the probability

Exercise: Conditional Probability

Exercises
Calculate the probability of each conditional event. Enter the result as a fraction. Use the / sign to indicate division. Always
reduce fractions to lowest terms.
1. If there are 6
red shirts and 7
green shirts, what is the probability that a red shirt is chosen and then another red shirt? These shirts are chosen without
replacement.
2. In the above situation, what is the probability that a red shirt is chosen and then a green shirt? These shirts are chosen
without replacement.
3. Again using the same clothing, what is the probability that a green shirt is chosen first and then a red shirt? These shirts
are chosen without replacement.
4. Using the same clothing, what is the probability a green shirt, and then another green shirt is chosen? These shirts are
chosen without replacement.
Use the information below to answer questions 5 - 7:

In 2016
approximately 41 %
of the population contributed to an employee-sponsored 401(k). Approximately 79 %
of the population works for an employer that sponsors a 401(k) plan. You are asked to find out the probability of someone
contributing to an employee-sponsored 401(k), given that he or she works for an employer that sponsors a 401(k) plan.

5. If the situation described above is represented atP(B | A)

, what are the events A
and B
?

a. A =
being offered a 401(k) through work; B =
contributing to 401(k) through work
b. A =
contributing to 401(k) through work; B =
being offered a 401(k) through work
c. A =
the probability of being offered a 401(k) through work; B =
the probability of contributing to 401(k) through work
d. A =
the probability of contributing to 401(k) through work; B =
the probability of being offered a 401(k) through work

6. Which of the following formulas should be used to solve forP(B | A)

?

a. P(B | A) = P(A) × P(B)

b. P(B | A) = P(A) + P(B)
c. P(B | A) = P(A and B) × P(A)
)

(|)
P(A and B)
d. P B A = P(A)

7. Using the information in questions 5 and 6, solve forP(B | A)

. Write the answer as a percent rounded to the nearest tenth. Do not enter the percent sign.

8. If you are asked to find the probability of either eventA

Copyright © 2019 MindEdge Inc. All rights reserved. Duplication prohibited.

8. If you are asked to find the probability of either eventA
happening or event B
happening, where events A
and B
are dependent which of the formulas below do you use?

a. P(A or B) = P(A) + P(B)

b. P(A and B) = P(A) + P(B) - P(A or B)
c. P(A and B) = P(A) × P(B | A)
d. P(A or B) = P(A) + P(B) - P(A and B)
9. If the probability that your favorite pitcher is pitching is50 %
, the probability that your baseball team will win is 50 %
, and the probability that your team will win and your favorite pitcher is pitching is30 %
, what is the probability that either your favorite team will win or your favorite pitcher will be pitching? (Enter your answer as a
decimal.)

10. It has been found that 81% of new cars sold in Notrealwood have a backup camera and59. 3 %
of cars sold have a 360-degree surround-view camera system. 52 %
of cars sold in Notrealwood have both. What is the probability that a car sold has had either a backup camera or a 360-
degree surround-view camera system? (Enter your answer as a percentage without the percentage sign.)

7.11.2 Determining Independence from Probabilities

Determining Independence from Probabilities

Thus far we have been told (or inferred) whether two events are independent. Now we will look at the probabilities of events to
determine whether they are independent or not.

Remember that P(A and B)

for two independent events is P(A) × P(B)
, while P(A and B)
for two dependent events is P(A) × P(B | A)
. It is worth determining whether two events are dependent or independent in order to determine whether you must find the
more complicated formula for P(A and B)
.

Determining Independence

Two Events are Independent If:

1. P(A and B) = P(A) · P(B)

2. P(A | B) = P(A)

3. P(B | A) = P(B)

4. P(B | A) = P(B | not A)

Please note, if any of these conditions holds, the events are independent

These conditions should make sense. The definition of independent events is that the occurrence of one does not affect the
probability of the other. So for formula 2 above [P(A | B) = P(A)]
, we are saying that it does not matter ifB
has occurred or not; A
still has the same chance of happening either way. Similarly, in formula 3 above, [P(B | A) = P(B)]
, we are saying that the probability ofB
does not change even if A

Copyright © 2019 MindEdge Inc. All rights reserved. Duplication prohibited.

has already happened. The third formula says this a different way, saying that the odds of B
happeneing are the same if A
has happened or if A
has not happened.

The first formula on the list P(A and B)

is defined as P(A) · P(B)
for independent events. If A
and B
are dependent events, remember the formula you should use is P(A and B) = P(A) · P(B | A)
.

Let us use a variation on our simple example to go through these equalities. LetA =
"flipping a head on the first flip," let B =
"flipping a head on the second flip." Unlike our earlier example, where B =
"flipping two heads," this B
is independent of A
. What you flip on the first coin toss has no effect on the outcome of the second coin toss. Using thisA
and B
, let us consider the equalities:

P(A and B) = P(A) · P(B)

The intersection of A
and B
is the toss HH. That is one toss out of four possible, so the

( )
1
P A and B = 4

( )
1
P A "tossing a heads on the first flip" = 2

( )
1
P A "tossing a heads on the second flip" = 2

() ()
1 1 1
P A ·P B = 2 · 2 = 4

Therefore:

P(A and B) = P(A) · P(B)

P(A | B) = P(A)

(|)
1
P A B = 2

There are two outcomes, TH and HH, that comprise event B

(tossing a H on the second throw). One of those outcomes, HH, is included in event A
(tossing a H on the first throw). Therefore the probability that A
occurs given B

Copyright © 2019 MindEdge Inc. All rights reserved. Duplication prohibited.

 1
has occurred is 2
2
. (Notice, this fraction is not 4
reduced. We are only considering the half of the sample space in which B
has occurred.)

()
1
P A = 2

There are two outcomes that comprise event A

()
1
: HT and HH. There are four total outcomes. ThereforeP A = 2

P(A | B) = P(A)

P(B | A) = P(B)

(|)
1

P B A = 2

The logic here is the same as that used to calculate P(A | B)

. There are two outcomes, HT and HH, that are included in eventA
(tossing a H on the first throw). One of those outcomes, HH, satisfied event B
(tossing a H on the second throw). Therefore the probability that B
occurs given A
1
has occurred is 2

()
1

P B = 2

There are two outcomes with which B

can occur, TH and HH; there are four outcomes total

P(B | A) = P(B)

P(B | A) = P(B | not A)

(|)
1

P B A = 2

(| )
1

P B not A = 2

P(not A)
has two possible outcomes, TH and TT. Of those two outcomes, only one satisfies event B
1
occurring, TH. Therefore the probability is 2

P(B | A) = P(B | not A)

Copyright © 2019 MindEdge Inc. All rights reserved. Duplication prohibited.

 Example
Eight toys are being given out to eight children in a pre-school class:

a green boat
a blue boat
a pink sparkly fairy doll
a blue sparkly fairy doll
a yellow toy telephone
a blue whale plush animal
a blue light-up star
a pink light-up star

1. Ari likes pink items and boats, so let A =

"getting a pink item" and B =
"getting a boat." Are
A and
B independent events?

Solution: checking if P(A | B) = P(A)

Let us select one of the equalities above to determine ifA

and B
are independent events. Calculate P(A | B)
and P(A)
.

P(A | B) = 0

B
is given, so Ari has gotten a boat. The boats don't come in pink, so the probability that Ari gets a pink item is 0
.

()
1
P A = 4

There are 2
pink items and 8
items total.

P(A | B) ≠ P(A), therefore the events are not independent.

2. Brianna likes blue items and boats. Let A =

"getting a blue item" and B =
"getting a boat" Are
A and
B independent events?

Solution: checking if P(A and B) = P(A) · P(B)

Copyright © 2019 MindEdge Inc. All rights reserved. Duplication prohibited.

 Let us use P(A and B) = P(A) · P(B)
to test for independence.

( )
1
P A and B = 8

There is one item that is both blue and a boat. The intersection of "getting a blue item" and "getting a boat" is the blue
boat.

()
1
P A = 2

There are 4
blue items and 8
items total.

()
1
P B = 4

There are 2
boats and 8
items altogether.

() ()
1 1 1
P A ·P B = 2 × 4 = 8

Therefore
P(A and B) = P(A) · P(B) and the events are independent

3. Leni likes pink items and fairy dolls. LetA =

"getting a pink item" and let B =
"getting a fairy doll" Are
A and
B independent events?

Solution: checking if P(B | A) = P(B | not A)

Let us use P(B | A) = P(B | not A)

to test for independence.

(|)
1
P B A = 2

Leni has gotten a pink item, now what is the probability of getting a fairy doll? There are 2
pink items and one is a fairy doll, so the probability is 0. 5
that she will get a fairy doll, given that she has gotten a pink item.

(| )
1
P B not A = 6

Copyright © 2019 MindEdge Inc. All rights reserved. Duplication prohibited.

 Leni has NOT gotten a pink item. There are 6
items that are not pink. One of those 6
items is a fairy doll. Therefore, the probability of B
given not A
1
is 6

Therefore,
P(B | not A) ≠ P(B | A) and the events
A and
B are not independent

7.12 Frequency Table Example

Frequency Table Example

We can now use two-way tables to examine important kinds of probability. For example, a potential home buyer might wonder
about the historical probability of a home gaining value in their area. Or a woman in her 30
s who is considering delaying having children might wonder what the probability if she can have two healthy children once she
is above 40
years old. Many important probabilities change according to other conditions.

In the table below, we consider a hypothetical study that tracked1, 000

homes according to their location and whether the home increased in value over the last 10 years or not.

Neighborhood and Home Value

Neighborhood:
Bleakburn Skystead Garmsby Totals
Home Value Status Increased Value 808 127 29 964
No Increase in Value 2 25 9 36
Totals 810 152 38 1000

To find probabilities:

1. Create events A
and B
based on the givens of the problem
2. Establish what kind of probability you will be finding, such as P(A and B)
or P(A | B)
depending on what the problem is asking
3. If necessary, determine if A
and B
are dependent or independent events
4. Write out the appropriate formula
5. Determine the probabilities needed for the formula (or take them from the givens of the problem)
6. Calculate the probability

Probabilities Using A Two-Way Table

Find the probability of a home in Bleakburn not increasing in value over the last 10 years.

Copyright © 2019 MindEdge Inc. All rights reserved. Duplication prohibited.

Solution

The intuitive way to do this problem is to notice that there are810

Bleakburn homes in the study and then to realize that only 2
of them did not increase in value.

P(not increasing value given home is in Bleakburn) =

2 1
810 = 405

Let us verify this by using our formula for conditional probability:

Step 1: Let A =
"being in Bleakburn"; Let B =
"not increasing in value"

Step 2: Determine P(B | A)

(|)
P(A and B)
Step 3:P B A = P(A)

Step 4: Fill in the probabilities you know. To do this you must calculateP(A)
and P(A and B)

() ( )
810 2
P A = 1000 P A and B = 1000

Remember that there are 2

homes that are both in Blackburn and did not increase in value out of1, 000
properties

P(B | A) =
2 810
1000 1000
÷

If you keep the probabilities as fractions you can cancel the numerator and denominator after converting the division
to a multiplication problem:
2 1000
1000 × 810

Notice in the calculation above, the 1000

s will cancel, leaving a numerator of 2
and a denominator of 810
. This will always be the case.
2 1000
1000 × 810

In the future, use the easier method of using the totals for event A as the denominator of your probability fraction.

Of the homes in Skystead, what proportion did not increase in value?

Copyright © 2019 MindEdge Inc. All rights reserved. Duplication prohibited.

 Solution

Find the total of the homes that are in Skystead:

152

Find the homes of those in Skystead that did not increase in value: 25

Divide the number of homes that did not increase in value by the total number of homes in Skystead:
25
152 = 16. 45 %

Extrapolating from the table above, if a property is in Garmsby, what is the probability that it did not increase in
value over the last 10 years?

Solution

Find the total of the homes in Garmsby:38

Find the homes of those in Garmsby that did not increase in value: 9

Divide the number of homes that did not increase in value by the total number of homes in Garmsby:
9
38 = 23. 68 %

7.13 Probability Trees

Probability Trees

Remember that earlier we built tree diagrams to help sort out the sample space of an event. For example, we used a tree
structure to determine the eight ways of having three children.

Probability trees are similar in structure, but they record information along the stems of the tree as well as at the nodes.
Consider the following situation:

Suppose at a work function, the boss is handing out donuts. You're brave enough to ask for another one after waiting10
minutes, but you're not brave enough to tell your boss that your favorite flavor is strawberry. (Also, no one else is eating
donuts.) If there are five delectable strawberry frosted donuts and three chocolate frosted donuts, what is the probability that
you get two strawberry ones?

Click on the following slides to learn how to construct a probability tree

Slide 1

Copyright © 2019 MindEdge Inc. All rights reserved. Duplication prohibited.

 Begin with the initial condition, 5
strawberry and 3
chocolate donuts. Draw two lines from the initial condition, illustrating the two possibilities: one of picking a strawberry
donut and one of picking a chocolate donut.

Slide 2

Label each line with the probability of that event. There is a5

in 8
possibility of randomly selecting the strawberry donut and a 3
in 8
possibility of randomly selecting a chocolate donut.

Slide 3

Copyright © 2019 MindEdge Inc. All rights reserved. Duplication prohibited.

 Now draw (or just write) the results at the end of the line. If the boss has selected a strawberry donut, there are now4
strawberry and 3
chocolate donuts. If the boss has selected a chocolate donut, there are now 2
chocolate and 5
strawberry donuts.

Slide 4

Draw in the next set of options and write in the probabilities above the lines.

Slide 5

Copyright © 2019 MindEdge Inc. All rights reserved. Duplication prohibited.

 You now have a probability tree. You can use the probability tree to calculate the probability of each of the branches.

Using Probability Trees

You can use the tree you have built to calculate conditional probabilities. To do this, calculate the probabilities for each branch
of the tree and then multiply the sequence of events you're interested in finding the probability for.

For example, what is the probability that your boss gives you a strawberry donut followed by a second strawberry donut? To do
this problem, multiply along the branch of the tree that involves selecting two strawberry donuts.

5 4 20 5
8 × 7 = 56 = 14 , or about
35. 7 % .

Why does this work? Recall the formula P(A and B) = P(A) × P(B | A)
. To find the probability of one branch of the tree you are solving forP(A and B)
—in this case strawberry and strawberry. The first branch gives you P(A)
(getting an initial strawberry donut) and the second gives you P(B | A)
(getting a second strawberry donut, given you already have one). In a probability tree, the formula can be extended
P(C | A and B)
etc. for each further branch.

You can use the same tree to calculate a different probability. What is the probability that your boss gives you a chocolate
donut followed by a strawberry donut?
3 5 15
8 × 7 = 56
or about 28. 8 %
.

Note that the probabilities of each line of the probability tree sum to1
. This is because the probability tree includes all possible scenarios, meaning that the sum of all the possibilities must add up
to 100 %
(or 1
) since all possibilities are accounted for.

Copyright © 2019 MindEdge Inc. All rights reserved. Duplication prohibited.

Notice that the probabilities of each branch sum to1
:
20 15 15 6 56
56 + 56 + 56 + 56 = 56

The fact that if you multiply across each branch and then add up the resulting products, you get1
is called the law of total probability. Even though the probabilities of individual sequences of events are not the same—some
are far more likely than others—the tree gives the complete set of probabilities for any sequence of events and therefore the
sum of the branches is 1
.

Example
A financial advisor is examining the performance of his clients' holdings to better inform his decisions. Given the following
probabilities, construct a probability tree to help determine certain probabilities:

60 %
of investments are in stocks.

40 %
of investments are in bonds.

72. 5 %
of stocks saw an annual gain.

15. 4 %
of bonds saw an annual loss.

1.) What is the probability that an investment is a stock that had an annual gain?

2.) What is the probability that an investment is a bond that had an annual gain?

3.) What is the probability that an investment is a stock that had an annual loss?

4.) What is the probability that an investment is a bond that had an annual loss?

Copyright © 2019 MindEdge Inc. All rights reserved. Duplication prohibited.

1.) What is the probability that an investment is a stock that had an annual gain?

Let A =
being a stock

Let B =
having an annual gain

P(being a stock that had an annual gain) = P(being a stock) × P(having an annual gain given being a stock investment)

P(A and B) = P(A) × P(B | A) = . 6 × 0. 725 = 0. 435

Interpretation:
43 - 44 investments out of
100 are stocks that have an annual gain.

2.) What is the probability that an investment is a bond that sees an annual gain?

Let A =
being a bond.

Let B =
having an annual gain.

P(being a bond that had an annual gain) = P(being a bond) × P(having an annual gain given being a bond investment)

P(A and B) = P(A) × P(B | A) = . 4 × 0. 846 = 0. 3384

Interpretation:
33 - 34 investments out of
100 are bonds that have an annual gain.

3.) What is the probability that an investment is a stock that had an annual loss?

Copyright © 2019 MindEdge Inc. All rights reserved. Duplication prohibited.

 Let A =
being a stock

Let B =
having an annual loss

P(being a stock that had an annual loss) = P(being a stock) × P(having an annual loss given being a stock investment)

P(A and B) = P(A) × P(B | A) = 0. 6 × 0. 275 = 0. 165

Interpretation:
16 - 17 investments out of
100 are stocks that have an annual loss.

4.) What is the probability that an investment is a bond that had an annual loss?

Let A =
being a bond

Let B =
having an annual loss

P(being a bond that had an annual loss) = P(being a bond) × P(having an annual loss given being a bond investment)

P(A and B) = P(A) × P(B | A) = . 4 × 0. 154 = 0. 0616

Interpretation: Around
6 investments out of
100 are bonds that have an annual loss.

Why are there more stocks with an annual gain than bonds with an annual gain? Because more investments
are in stocks, the prevalence of stocks with gains will be higher than bonds with gains, even though bonds
are more likely to have an annual gain.

Let us use the probability tree above to verify the law of total probability.

First, multiply each branch of the tree through:

Copyright © 2019 MindEdge Inc. All rights reserved. Duplication prohibited.

Now sum each product, moving down the outside of the tree:

7.14 Flashcards
Module 7 Flashcards

Term Definition

7.15 Review Game

7.19 Module 7 Review Test

7.20 Module Feedback

Copyright © 2019 MindEdge Inc. All rights reserved. Duplication prohibited.