Chapter 4 Basic Probability - 2 PDF

4 Basic Probability
USING STATISTICS @ M&R Decision Trees Counting Rule 3

Electronics World Independence Counting Rule 4
Multiplication Rules Counting Rule 5
4.1 Basic Probability Concepts Marginal Probability Using
Events and Sample Spaces the General Multiplication 4.5 Ethical Issues and
Contingency Tables and Rule Probability
Venn Diagrams
Simple Probability 4.3 Bayes Theorem USING STATISTICS @ M&R
Joint Probability Electronics World Revisited
Marginal Probability THINK ABOUT THIS: Divine
General Addition Rule Providence and Spam CHAPTER 4 EXCEL GUIDE
4.2 Conditional Probability 4.4 Counting Rules CHAPTER 4 MINITAB GUIDE

Computing Conditional Counting Rule 1
Probabilities Counting Rule 2
Learning Objectives
In this chapter, you learn:
Basic probability concepts

Conditional probability
Bayes theorem to revise probabilities
Various counting rules
U S I N G S TAT I S T I C S
@ M&R Electronics World
A s the marketing manager for M&R Electronics World, you are analyzing the survey
results of an intent-to-purchase study. This study asked the heads of 1,000 households
about their intentions to purchase a big-screen television sometime during the next
12 months. As a follow-up, you plan to survey the same people 12 months later to see
whether they purchased televisions. In addition, for households purchasing big-screen
televisions, you would like to know whether the television they purchased had a faster refresh rate (120 Hz
or higher) or a standard refresh rate (60 Hz), whether they also purchased a Blu-ray disc (BD) player in the
past 12 months, and whether they were satisfied with their purchase of the big-screen television.
You are expected to use the results of this survey to plan a new marketing strategy that will enhance
sales and better target those households likely to purchase multiple or more expensive products. What
questions can you ask in this survey? How can you express the relationships among the various intent-
to-purchase responses of individual households?
In previous chapters, you learned descriptive methods to summarize categorical and numerical
variables. In this chapter, you will learn about probability to answer questions such as the following:
What is the probability that a household is planning to purchase a
big-screen television in the next year?
What is the probability that a household will actually purchase
a big-screen television?
What is the probability that a household is planning to purchase a
big-screen television and actually purchases the television?
Given that the household is planning to purchase a big-screen televi-
sion, what is the probability that the purchase is made?
Does knowledge of whether a household plans to purchase the televi-
sion change the likelihood of predicting whether the household will
purchase the television?
What is the probability that a household that purchases a big-screen
television will purchase a television with a faster refresh rate?
television with a faster refresh rate will also purchase a Blu-ray disc
player?
television will be satisfied with the purchase?
With answers to questions such as these, you can begin to make deci-
sions about your marketing strategy. Should your strategy for selling
more big-screen televisions target those households that have indicated
an intent to purchase? Should you concentrate on selling televisions that
have faster refresh rates? Is it likely that households that purchase big-
screen televisions with faster refresh rates can be easily persuaded to also
purchase Blu-ray disc players? 145
146 CHAPTER 4 Basic Probability
he principles of probability help bridge the worlds of descriptive statistics and
T inferential statistics. Reading this chapter will help you learn about different types of
probabilities, how to compute probabilities, and how to revise probabilities in light of
new information. Probability principles are the foundation for the probability distribution, the
concept of mathematical expectation, and the binomial, Poisson, and hypergeometric distribu-
tions, topics that are discussed in Chapter 5.
4.1 Basic Probability Concepts

What is meant by the word probability? A probability is the numeric value representing the
chance, likelihood, or possibility that a particular event will occur, such as the price of a stock
increasing, a rainy day, a defective product, or the outcome five dots in a single toss of a die.
In all these instances, the probability involved is a proportion or fraction whose value ranges
between 0 and 1, inclusive. An event that has no chance of occurring (the impossible event)
has a probability of 0. An event that is sure to occur (the certain event) has a probability of 1.
There are three types of probability:
A priori
Empirical
Subjective
In a priori probability, the probability of an occurrence is based on prior knowledge of the
process involved. In the simplest case, where each outcome is equally likely, the chance of
occurrence of the event is defined in Equation (4.1).
PROBABILITY OF OCCURRENCE
X
Probability of occurrence = (4.1)
T
where
X = number of ways in which the event occurs

T = total number of possible outcomes
Consider a standard deck of cards that has 26 red cards and 26 black cards. The probabil-
ity of selecting a black card is 26>52 = 0.50 because there are X = 26 black cards and
T = 52 total cards. What does this probability mean? If each card is replaced after it is
selected, does it mean that 1 out of the next 2 cards selected will be black? No, because you
cannot say for certain what will happen on the next several selections. However, you can say
that in the long run, if this selection process is continually repeated, the proportion of black
cards selected will approach 0.50. Example 4.1 shows another example of computing an
a priori probability.
EXAMPLE 4.1 A standard six-sided die has six faces. Each face of the die contains either one, two, three, four,
five, or six dots. If you roll a die, what is the probability that you will get a face with five dots?
Finding A Priori
Probabilities SOLUTION Each face is equally likely to occur. Because there are six faces, the probability
of getting a face with five dots is 1/6.
4.1 Basic Probability Concepts 147
The preceding examples use the a priori probability approach because the number of ways the
event occurs and the total number of possible outcomes are known from the composition of
the deck of cards or the faces of the die.
In the empirical probability approach, the probabilities are based on observed data, not
on prior knowledge of a process. Surveys are often used to generate empirical probabilities.
Examples of this type of probability are the proportion of individuals in the Using Statistics
scenario who actually purchase big-screen televisions, the proportion of registered voters who
prefer a certain political candidate, and the proportion of students who have part-time jobs.
For example, if you take a survey of students, and 60% state that they have part-time jobs, then
there is a 0.60 probability that an individual student has a part-time job.
The third approach to probability, subjective probability, differs from the other two
approaches because subjective probability differs from person to person. For example, the
development team for a new product may assign a probability of 0.60 to the chance of success
for the product, while the president of the company may be less optimistic and assign a proba-
bility of 0.30. The assignment of subjective probabilities to various outcomes is usually based
on a combination of an individuals past experience, personal opinion, and analysis of a partic-
ular situation. Subjective probability is especially useful in making decisions in situations in
which you cannot use a priori probability or empirical probability.
Events and Sample Spaces

The basic elements of probability theory are the individual outcomes of a variable under study.
You need the following definitions to understand probabilities.
EVENT
Each possible outcome of a variable is referred to as an event.
A simple event is described by a single characteristic.
For example, when you toss a coin, the two possible outcomes are heads and tails. Each of
these represents a simple event. When you roll a standard six-sided die in which the six faces
of the die contain either one, two, three, four, five, or six dots, there are six possible simple
events. An event can be any one of these simple events, a set of them, or a subset of all of them.
For example, the event of an even number of dots consists of three simple events (i.e., two,
four, or six dots).
JOINT EVENT
A joint event is an event that has two or more characteristics.
Getting two heads when you toss a coin twice is an example of a joint event because it consists
of heads on the first toss and heads on the second toss.
COMPLEMENT
The complement of event A (represented by the symbol A ) includes all events that are
not part of A.
The complement of a head is a tail because that is the only event that is not a head. The com-
plement of five dots on a die is not getting five dots. Not getting five dots consists of getting
one, two, three, four, or six dots.
SAMPLE SPACE
The collection of all the possible events is called the sample space.
The sample space for tossing a coin consists of heads and tails. The sample space
when rolling a die consists of one, two, three, four, five, and six dots. Example 4.2
demonstrates events and sample spaces.
EXAMPLE 4.2 The Using Statistics scenario on page 145 concerns M&R Electronics World. Table 4.1 pres-
ents the results of the sample of 1,000 households in terms of purchase behavior for big-screen
Events and Sample televisions.
Spaces
TA B L E 4 . 1 ACTUALLY PURCHASED
PLANNED
Purchase Behavior for TO PURCHASE Yes No Total
Big-Screen Televisions
Yes 200 50 250
No 100 650 750
Total 300 700 1,000
What is the sample space? Give examples of simple events and joint events.
SOLUTION The sample space consists of the 1,000 respondents. Simple events are planned
to purchase, did not plan to purchase, purchased, and did not purchase. The comple-
ment of the event planned to purchase is did not plan to purchase. The event planned to
purchase and actually purchased is a joint event because in this joint event the respondent
must plan to purchase the television and actually purchase it.
Contingency Tables and Venn Diagrams

There are several ways in which you can view a particular sample space. One way involves
using a contingency table (see Section 2.2) such as the one displayed in Table 4.1. You get the
values in the cells of the table by subdividing the sample space of 1,000 households according
to whether someone planned to purchase and actually purchased a big-screen television set.
For example, 200 of the respondents planned to purchase a big-screen television set and subse-
quently did purchase the big-screen television set.
A second way to present the sample space is by using a Venn diagram. This diagram graph-
ically represents the various events as unions and intersections of circles. Figure 4.1 presents
a typical Venn diagram for a two-variable situation, with each variable having only two events
(A and A , B and B ). The circle on the left (the red one) represents all events that are part of A.
The circle on the right (the yellow one) represents all events that are part of B. The area
contained within circle A and circle B (center area) is the intersection of A and B (written as
A B), since it is part of A and also part of B. The total area of the two circles is the union of
A and B (written as A B) and contains all outcomes that are just part of event A, just part of
event B, or part of both A and B. The area in the diagram outside of A B contains outcomes
that are neither part of A nor part of B.
You must define A and B in order to develop a Venn diagram. You can define either event
as A or B, as long as you are consistent in evaluating the various events. For the big-screen
television example, you can define the events as follows:
A = planned to purchase B = actually purchased
A = did not plan to purchase B = did not actually purchase
In drawing the Venn diagram (see Figure 4.2), you must determine the value of the inter-
section of A and B so that the sample space can be divided into its parts. A B consists of all
200 households who planned to purchase and actually purchased a big-screen television set.
FIGURE 4.1 B FIGURE 4.2 A B A B = 650

A B
Venn diagram for events Venn diagram for the
A and B M&R Electronics World
A B
example
50 200 100
A
A B
A B = 350
The remainder of event A (planned to purchase) consists of the 50 households who planned to
purchase a big-screen television set but did not actually purchase one. The remainder of event
B (actually purchased) consists of the 100 households who did not plan to purchase a big-
screen television set but actually purchased one. The remaining 650 households represent
those who neither planned to purchase nor actually purchased a big-screen television set.
Simple Probability
Now you can answer some of the questions posed in the Using Statistics scenario. Because the
results are based on data collected in a survey (refer to Table 4.1), you can use the empirical
probability approach.
As stated previously, the most fundamental rule for probabilities is that they range in value
from 0 to 1. An impossible event has a probability of 0, and an event that is certain to occur
has a probability of 1.
Simple probability refers to the probability of occurrence of a simple event, P(A).
A simple probability in the Using Statistics scenario is the probability of planning to purchase
a big-screen television. How can you determine the probability of selecting a household that
planned to purchase a big-screen television? Using Equation (4.1) on page 146:
X
Probability of occurrence =
T
Number who planned to purchase
P(Planned to purchase) =
Total number of households
250
= = 0.25
1,000
Thus, there is a 0.25 (or 25%) chance that a household planned to purchase a big-screen television.
Example 4.3 illustrates another application of simple probability.
EXAMPLE 4.3 In the Using Statistics follow-up survey, additional questions were asked of the 300 households
that actually purchased big-screen televisions. Table 4.2 indicates the consumers responses to
Computing the whether the television purchased had a faster refresh rate and whether they also purchased a
Probability That the Blu-ray disc (BD) player in the past 12 months.
Big-Screen Television Find the probability that if a household that purchased a big-screen television is randomly
Purchased Had a selected, the television purchased had a faster refresh rate.
Faster Refresh Rate
TA B L E 4 . 2 PURCHASED BD PLAYER
REFRESH RATE OF
Purchase Behavior TELEVISION PURCHASED Yes No Total
Regarding Purchasing
a Faster Refresh Rate Faster 38 42 80
Television and Blu-Ray Standard 70 150 220
Disc (BD) Player Total 108 192 300
SOLUTION Using the following definitions:

A = purchased a television with a faster refresh rate
A = purchased a television with a standard refresh rate
B = purchased a Blu-ray disc (BD) player
B = did not purchase a Blu-ray disc (BD) player
Number of faster refresh rate televisions
P(faster refresh rate) =
Total number of televisions
80
= = 0.267
300
There is a 26.7% chance that a randomly selected big-screen television purchased has a faster
refresh rate.
Joint Probability
Whereas simple or marginal probability refers to the probability of occurrence of simple
events, joint probability refers to the probability of an occurrence involving two or more
events. An example of joint probability is the probability that you will get heads on the first
toss of a coin and heads on the second toss of a coin.
In Table 4.1 on page 148, the group of individuals who planned to purchase and actually
purchased a big-screen television consist only of the outcomes in the single cell yesplanned
to purchase and yesactually purchased. Because this group consists of 200 households, the
probability of picking a household that planned to purchase and actually purchased a big-
screen television is
Planned to purchase and actually purchased

P(Planned to purchase and actually purchased) =
Total number of respondents
200
= = 0.20
1,000
Example 4.4 also demonstrates how to determine joint probability.
EXAMPLE 4.4 In Table 4.2, the purchases are cross-classified as having a faster refresh rate or having a stan-
dard refresh rate and whether the household purchased a Blu-ray disc player. Find the proba-
Determining the bility that a randomly selected household that purchased a big-screen television also purchased
Joint Probability a television that had a faster refresh rate and purchased a Blu-ray disc player.
That a Household
SOLUTION Using Equation (4.1) on page 146,
Purchased a Big-
Screen Television Number that purchased a television with a faster
P(television with a faster refresh refresh rate and a Blu-ray disc player
with a Faster =
rate and Blu-ray disc player)
Refresh Rate and Total number of big-screen television purchasers
a Blu-ray Disc Player 38
= = 0.127
300
Therefore, there is a 12.7% chance that a randomly selected household that purchased a big-
screen television purchased a television that had a faster refresh rate and a Blu-ray disc player.
Marginal Probability
The marginal probability of an event consists of a set of joint probabilities. You can determine
the marginal probability of a particular event by using the concept of joint probability just dis-
cussed. For example, if B consists of two events, B1 and B2, then P(A), the probability of event A,
consists of the joint probability of event A occurring with event B1 and the joint probability of
event A occurring with event B2. You use Equation (4.2) to compute marginal probabilities.
MARGINAL PROBABILITY
P(A) = P(A and B1) + P(A and B2) + + P(A and Bk) (4.2)
where B1, B2, . . . , Bk are k mutually exclusive and collectively exhaustive events, defined
as follows:
Two events are mutually exclusive if both the events cannot occur simultaneously.
A set of events is collectively exhaustive if one of the events must occur.
Heads and tails in a coin toss are mutually exclusive events. The result of a coin toss cannot
simultaneously be a head and a tail. Heads and tails in a coin toss are also collectively exhaus-
tive events. One of them must occur. If heads does not occur, tails must occur. If tails does not
occur, heads must occur. Being male and being female are mutually exclusive and collectively
exhaustive events. No person is both (the two are mutually exclusive), and everyone is one or
the other (the two are collectively exhaustive).
You can use Equation (4.2) to compute the marginal probability of planned to purchase
a big-screen television:
P(Planned to purchase) = P(Planned to purchase and purchased)

+ P(Planned to purchase and did not purchase)
200 50
= +
1,000 1,000
250
= = 0.25
1,000
You get the same result if you add the number of outcomes that make up the simple event
planned to purchase.
General Addition Rule

How do you find the probability of event A or B? You need to consider the occurrence of either
event A or event B or both A and B. For example, how can you determine the probability that a
household planned to purchase or actually purchased a big-screen television? The event planned
to purchase or actually purchased includes all households that planned to purchase and all
households that actually purchased a big-screen television. You examine each cell of the contin-
gency table (Table 4.1 on page 148) to determine whether it is part of this event. From Table 4.1,
the cell planned to purchase and did not actually purchase is part of the event because it in-
cludes respondents who planned to purchase. The cell did not plan to purchase and actually pur-
chased is included because it contains respondents who actually purchased. Finally, the cell
planned to purchase and actually purchased has both characteristics of interest. Therefore, one
way to calculate the probability of planned to purchase or actually purchased is
P(Planned to purchase or actually purchased) = P(Planned to purchase and did

not actually purchase) + P(Did not plan to
purchase and actually purchase) + P(Planned
to purchase and actually purchased)
50 100 200
= + +
1,000 1,000 1,000
350
= = 0.35
1,000
Often, it is easier to determine P(A or B), the probability of the event A or B, by using the
general addition rule, defined in Equation (4.3).
GENERAL ADDITION RULE

The probability of A or B is equal to the probability of A plus the probability of B minus
the probability of A and B.
P(A or B) = P(A) + P(B) - P(A and B) (4.3)
Applying Equation (4.3) to the previous example produces the following result:
P(Planned to purchase or actually purchased) = P(Planned to purchase)

+ P(Actually purchased) - P(Planned to
purchase and actually purchased)
250 300 200
= + -
1,000 1,000 1,000
350
= = 0.35
1,000
The general addition rule consists of taking the probability of A and adding it to the
probability of B and then subtracting the probability of the joint event A and B from this
total because the joint event has already been included in computing both the probability of
A and the probability of B. Referring to Table 4.1 on page 148, if the outcomes of the event
planned to purchase are added to those of the event actually purchased, the joint event
planned to purchase and actually purchased has been included in each of these simple
events. Therefore, because this joint event has been double-counted, you must subtract it to
provide the correct result. Example 4.5 illustrates another application of the general
addition rule.
EXAMPLE 4.5 In Example 4.3 on page 149, the purchases were cross-classified in Table 4.2 as televisions
that had a faster refresh rate or televisions that had a standard refresh rate and whether the
Using the General household purchased a Blu-ray disc (BD) player. Find the probability that among households
Addition Rule for that purchased a big-screen television, they purchased a television that had a faster refresh rate
the Households or a BD player.
That Purchased
SOLUTION Using Equation (4.3),
Big-Screen
Televisions P(Television had a faster refresh rate)
P(Television had a faster refresh = + P(purchased a BD player) P(Television
-
rate or purchased a BD player) had a faster refresh rate and purchased a BD player)
80 108 38
= + -
300 300 300
150
= = 0.50
300
Therefore, of those households that purchased a big-screen television, there is a 50.0% chance
that a randomly selected household purchased a television that had a faster refresh rate or
purchased a BD player.
Problems for Sections 4.1 153
Problems for Section 4.1

LEARNING THE BASICS b. Each respondent was classified by the type of car he or she
4.1 Two coins are tossed. drives: sedan, SUV, American, European, Asian, or none.
a. Give an example of a simple event. c. People were asked, Do you currently live in (i) an apart-
b. Give an example of a joint event. ment or (ii) a house?
c. What is the complement of a head on the first toss? d. A product was classified as defective or not defective.
4.2 An urn contains 12 red balls and 8 white balls. One ball 4.7 Which of the following events occur with a probability
is to be selected from the urn. of zero? For each, state why or why not.
a. Give an example of a simple event. a. A voter in the United States is registered as a Republican
b. What is the complement of a red ball? and as a Democrat.
b. A voter in the United States is female and registered as a
4.3 Consider the following contingency table: Republican.
c. An automobile is a Ford and a Toyota.
B B d. An automobile is a Toyota and was manufactured in the
United States.
A 10 20
A 20 40 4.8 Does it take more time to be removed from an email
list than it used to take? A study of 100 large online retailers
revealed the following:
What is the probability of
a. event A?
b. event A? NEED THREE OR MORE CLICKS TO BE REMOVED
c. event A and B? YEAR Yes No
d. A or B?
2009 39 61
4.4 Consider the following contingency table: 2008 7 93
Source: Data extracted from More Clicks to Escape an Email List,
B B The New York Times, March 29, 2010, p. B2.
A 10 30
A 25 35 a. Give an example of a simple event.
b. Give an example of a joint event.
c. What is the complement of Needs three or more clicks
What is the probability of
to be removed from an email list?
a. event A?
d. Why is Needs three or more clicks to be removed from
b. event A and B?
an email list in 2009 a joint event?
c. event A and B?
d. event A or B? 4.9 Referring to the contingency table in Problem 4.8, if a
large online retailer is selected at random, what is the prob-
APPLYING THE CONCEPTS ability that
a. you needed three or more clicks to be removed from an
4.5 For each of the following, indicate whether the type of email list?
probability involved is an example of a priori probability, b. you needed three or more clicks to be removed from an
empirical probability, or subjective probability. email list in 2009?
a. The next toss of a fair coin will land on heads. c. you needed three or more clicks to be removed from an
b. Italy will win soccers World Cup the next time the com- email list or were a large online retailer surveyed in 2009?
petition is held. d. Explain the difference in the results in (b) and (c).
c. The sum of the faces of two dice will be seven.
4.10 Do people of different age groups differ in their
d. The train taking a commuter to work will be more than
response to email messages? A survey by the Center for the
10 minutes late.
Digital Future of the University of Southern California (data
4.6 For each of the following, state whether the events extracted from A. Mindlin, Older E-mail Users Favor Fast
created are mutually exclusive and collectively exhaustive. Replies, The New York Times, July 14, 2008, p. B3) re-
a. Registered voters in the United States were asked whether ported that 70.7% of users over 70 years of age believe that
they are registered as Republicans or Democrats. email messages should be answered quickly, as compared to
53.6% of users 12 to 50 years old. Suppose that the survey 4.13 What is the preferred way for people to order fast
was based on 1,000 users over 70 years of age and 1,000 food? A survey was conducted in 2009, but the sample sizes
users 12 to 50 years old. The following table summarizes the were not reported. Suppose the results, based on a sample of
results: 100 males and 100 females, were as follows:
AGE OF RESPONDENTS GENDER

ANSWERS QUICKLY 1250 Over 70 Total DINING PREFERENCE Male Female Total
Yes 536 707 1,243 Dine inside 21 12 33
No 464 293 757 Order inside to go 19 10 29
Total 1,000 1,000 2,000 Order at the
drive-through 60 78 138
Total 100 100 200
a. Give an example of a simple event. Source: Data extracted from www.qsrmagazine.com/reports/drive-
b. Give an example of a joint event. thru_time_study/2009/2009_charts/whats_your_preferred_way_
c. What is the complement of a respondent who answers to_order_fast_food.html.
quickly?
d. Why is a respondent who answers quickly and is over 70
If a respondent is selected at random, what is the probability
years old a joint event?
that he or she
4.11 Referring to the contingency table in Problem 4.10, if a. prefers to order at the drive-through?
a respondent is selected at random, what is the probability b. is a male and prefers to order at the drive-through?
that c. is a male or prefers to order at the drive-through?
a. he or she answers quickly? d. Explain the difference in the results in (b) and (c).
b. he or she is over 70 years old?
4.14 A sample of 500 respondents in a large metropolitan
c. he or she answers quickly or is over 70 years old?
area was selected to study consumer behavior. Among the
d. Explain the difference in the results in (b) and (c).
questions asked was Do you enjoy shopping for clothing?
SELF 4.12 According to a Gallup Poll, the extent to Of 240 males, 136 answered yes. Of 260 females, 224
Test which employees are engaged with their work- answered yes. Construct a contingency table to evaluate the
place varies from country to country. Gallup reports that probabilities. What is the probability that a respondent
the percentage of U.S. workers engaged with their work- chosen at random
place is more than twice as high as the percentage of a. enjoys shopping for clothing?
German workers. The study also shows that having more b. is a female and enjoys shopping for clothing?
engaged workers leads to increased innovation, productiv- c. is a female or enjoys shopping for clothing?
ity, and profitability, as well as reduced employee turnover. d. is a male or a female?
The results of the poll are summarized in the following
4.15 Each year, ratings are compiled concerning the
table:
performance of new cars during the first 90 days of use.
Suppose that the cars have been categorized according to
COUNTRY whether a car needs warranty-related repair (yes or no) and
the country in which the company manufacturing a car is
ENGAGEMENT United States Germany Total based (United States or not United States). Based on the data
Engaged 550 246 796 collected, the probability that the new car needs a warranty
Not engaged 1,345 1,649 2,994 repair is 0.04, the probability that the car was manufactured
Total 1,895 1,895 3,790 by a U.S.-based company is 0.60, and the probability that the
new car needs a warranty repair and was manufactured by a
Source: Data extracted from M. Nink, Employee Disengagement
Plagues Germany, Gallup Management Journal, gmj.gallup.com,
U.S.-based company is 0.025. Construct a contingency table
April 9, 2009. to evaluate the probabilities of a warranty-related repair.
What is the probability that a new car selected at random
a. needs a warranty repair?
If an employee is selected at random, what is the probability b. needs a warranty repair and was manufactured by a U.S.-
that he or she based company?
a. is engaged with his or her workplace? c. needs a warranty repair or was manufactured by a U.S.-
b. is a U.S. worker? based company?
c. is engaged with his or her workplace or is a U.S. worker? d. needs a warranty repair or was not manufactured by a
d. Explain the difference in the results in (b) and (c). U.S.-based company?
4.2 Conditional Probability 155
4.2 Conditional Probability

Each example in Section 4.1 involves finding the probability of an event when sampling from
the entire sample space. How do you determine the probability of an event if you know certain
information about the events involved?
Computing Conditional Probabilities

Conditional probability refers to the probability of event A, given information about the
occurrence of another event, B.
CONDITIONAL PROBABILITY
The probability of A given B is equal to the probability of A and B divided by the
probability of B.
P1A and B2
P1A|B2 = (4.4a)
P1B2
The probability of B given A is equal to the probability of A and B divided by the
probability of A.
P1A and B2
P1B|A2 = (4.4b)
P1A2
where
P(A and B) = joint probability of A and B
P(A) = marginal probability of A
P(B) = marginal probability of B
Referring to the Using Statistics scenario involving the purchase of big-screen televi-
sions, suppose you were told that a household planned to purchase a big-screen television.
Now, what is the probability that the household actually purchased the television? In this ex-
ample, the objective is to find P(Actually purchased | Planned to purchase). Here you are
given the information that the household planned to purchase the big-screen television.
Therefore, the sample space does not consist of all 1,000 households in the survey. It consists
of only those households that planned to purchase the big-screen television. Of 250 such
households, 200 actually purchased the big-screen television. Therefore, based on Table 4.1
on page 148, the probability that a household actually purchased the big-screen television
given that he or she planned to purchase is
Planned to purchase and actually purchased

P1Actually purchased Planned to purchase2 =
Planned to purchase
200
= = 0.80
250
You can also use Equation (4.4b) to compute this result:
P1A and B2
P1B|A2 =
P1A2
where
A = planned to purchase
B = actually purchased
then
200>1,000
250>1,000
200
= = 0.80
250
Example 4.6 further illustrates conditional probability.
EXAMPLE 4.6 Table 4.2 on page 149 is a contingency table for whether a household purchased a television
with a faster refresh rate and whether the household purchased a Blu-ray disc player. If a
Finding the Condi- household purchased a television with a faster refresh rate, what is the probability that it also
tional Probability purchased a Blu-ray disc player?
of Purchasing
SOLUTION Because you know that the household purchased a television with a faster
a Blu-ray Disc
refresh rate, the sample space is reduced to 80 households. Of these 80 households, 38 also
Player purchased a Blu-ray disc (BD) player. Therefore, the probability that a household
purchased a BD player, given that the household purchased a television with a faster refresh
rate, is
Number purchasing television with

P1Purchased BD player | Purchased faster refresh rate and BD player
television with faster refresh rate) = Number purchasing television
with faster refresh rate
38
= = 0.475
80
If you use Equation (4.4b) on page 155:

A = purchased a television with a faster refresh rate
B = purchased a BD player
then
P1A and B2 38>300

P1B|A2 = = = 0.475
P1A2 80>300
Therefore, given that the household purchased a television with a faster refresh rate,
there is a 47.5% chance that the household also purchased a Blu-ray disc player. You can
compare this conditional probability to the marginal probability of purchasing a Blu-ray
disc player, which is 108>300 = 0.36, or 36%. These results tell you that households that
purchased televisions with a faster refresh rate are more likely to purchase a Blu-ray disc
player than are households that purchased big-screen televisions that have a standard
refresh rate.
Decision Trees
In Table 4.1 on page 148, households are classified according to whether they planned to
purchase and whether they actually purchased big-screen televisions. A decision tree is an
alternative to the contingency table. Figure 4.3 represents the decision tree for this
example.
FIGURE 4.3
ed P(A and B) = 200
Decision tree for M&R has
P(A) =
250 lly Purc 1,000
Electronics World 1,000 Actua
example
to
ned Did N
Plan hase ot A 50
Purc Purch ctually P(A and B)=
Entire ase 1,000
Set of
Households
Did
to P Not Pl hased
urc a
has n y Purc P(A and B) = 100
e Actuall 1,000
750
P(A) Did
1,000 Not
Pur Actua
cha l P(A and B) = 650
se ly 1,000
In Figure 4.3, beginning at the left with the entire set of households, there are two
branches for whether or not the household planned to purchase a big-screen television. Each
of these branches has two subbranches, corresponding to whether the household actually pur-
chased or did not actually purchase the big-screen television. The probabilities at the end of
the initial branches represent the marginal probabilities of A and A. The probabilities at the
end of each of the four subbranches represent the joint probability for each combination of
events A and B. You compute the conditional probability by dividing the joint probability by
the appropriate marginal probability.
For example, to compute the probability that the household actually purchased, given that
the household planned to purchase the big-screen television, you take P(Planned to purchase
and actually purchased) and divide by P(Planned to purchase). From Figure 4.3,
200>1,000
250>1,000
200
= = 0.80
250
Example 4.7 illustrates how to construct a decision tree.
EXAMPLE 4.7 Using the cross-classified data in Table 4.2 on page 149, construct the decision tree. Use the
decision tree to find the probability that a household purchased a Blu-ray disc player, given
Constructing the that the household purchased a television with a faster refresh rate.
Decision Tree for
the Households SOLUTION The decision tree for purchased a Blu-ray disc player and a television with a
faster refresh rate is displayed in Figure 4.4 on page 156. Using Equation (4.4b) on page 155
That Purchased
and the following definitions,
Big-Screen
Televisions A = purchased a television with a faster refresh rate
B = purchased a Blu-ray disc player
P1A and B2 38>300
P1B|A2 = = = 0.475
P1A2 80>300
FIGURE 4.4
r 38
Decision tree for aye
80 B D Pl P(A and B) =
300
purchased a television P(A) = h ased
300 Purc
with a faster refresh rate
and a Blu-ray disc (BD) er
Fast n
player h a sed levisio Did N
ot Pu
c e
Pur Rate T BD P rchase P(A and B) = 42
Entire esh layer 300
Set of Refr
Households Did
Fas Not P
ter
Re urch Player P(A and B) = 70
Tele fresh ase sed BD
visi R Purcha 300
on ate
220
P(A) = Did
300 Not
BD Purch
Pla a
yer se P(A and B) = 150
300
Independence
In the example concerning the purchase of big-screen televisions, the conditional probability
is 200>250 = 0.80 that the selected household actually purchased the big-screen television,
given that the household planned to purchase. The simple probability of selecting a household
that actually purchased is 300>1,000 = 0.30. This result shows that the prior knowledge that
the household planned to purchase affected the probability that the household actually
purchased the television. In other words, the outcome of one event is dependent on the
outcome of a second event.
When the outcome of one event does not affect the probability of occurrence of another
event, the events are said to be independent. Independence can be determined by using
Equation (4.5).
INDEPENDENCE
Two events, A and B, are independent if and only if
P1A|B2 = P1A2 (4.5)
where
P1A|B2 = conditional probability of A given B
P1A2 = marginal probability of A
Example 4.8 demonstrates the use of Equation (4.5).
EXAMPLE 4.8 In the follow-up survey of the 300 households that actually purchased big-screen televisions,
the households were asked if they were satisfied with their purchases. Table 4.3 cross-classi-
Determining fies the responses to the satisfaction question with the responses to whether the television had
Independence a faster refresh rate.
TA B L E 4 . 3 SATISFIED WITH PURCHASE?

TELEVISION
Satisfaction with REFRESH RATE Yes No Total
Purchase of Big-Screen
Televisions Faster 64 16 80
Standard 176 44 220
Total 240 60 300
Determine whether being satisfied with the purchase and the refresh rate of the television
purchased are independent.
SOLUTION For these data,
64>300 64
P1Satisfied faster refresh rate2 = = = 0.80
80>300 80
which is equal to
240
P1Satisfied2 = = 0.80
300
Thus, being satisfied with the purchase and the refresh rate of the television purchased are
independent. Knowledge of one event does not affect the probability of the other event.
Multiplication Rules
The general multiplication rule is derived using Equation (4.4a) on page 155:
P1A and B2
P1A|B2 =
P1B2
and solving for the joint probability P(A and B).
GENERAL MULTIPLICATION RULE

The probability of A and B is equal to the probability of A given B times the probability
of B.
P1A and B2 = P1A|B2P1B2 (4.6)
Example 4.9 demonstrates the use of the general multiplication rule.
EXAMPLE 4.9 Consider the 80 households that purchased televisions that had a faster refresh rate. In Table
4.3 on page 158 you see that 64 households are satisfied with their purchase, and 16 house-
Using the General holds are dissatisfied. Suppose 2 households are randomly selected from the 80 households.
Multiplication Rule Find the probability that both households are satisfied with their purchase.
SOLUTION Here you can use the multiplication rule in the following way. If
A = second household selected is satisfied
B = first household selected is satisfied
then, using Equation (4.6),
P1A and B2 = P1A|B2P1B2
The probability that the first household is satisfied with the purchase is 64/80. However,
the probability that the second household is also satisfied with the purchase depends on the
result of the first selection. If the first household is not returned to the sample after the satis-
faction level is determined (i.e., sampling without replacement), the number of households
remaining is 79. If the first household is satisfied, the probability that the second is also
satisfied is 63/79 because 63 satisfied households remain in the sample. Therefore,
P1A and B2 = a b a b = 0.6380

63 64
79 80
There is a 63.80% chance that both of the households sampled will be satisfied with their
purchase.
The multiplication rule for independent events is derived by substituting P(A) for
P1A|B2 in Equation (4.6).
MULTIPLICATION RULE FOR INDEPENDENT EVENTS

If A and B are independent, the probability of A and B is equal to the probability of A
times the probability of B.
P1A and B2 = P1A2P1B2 (4.7)
If this rule holds for two events, A and B, then A and B are independent. Therefore, there are
two ways to determine independence:
1. Events A and B are independent if, and only if, P1A|B2 = P1A2.
2. Events A and B are independent if, and only if, P1A and B2 = P1A2P1B2.
Marginal Probability Using the General Multiplication Rule

In Section 4.1, marginal probability was defined using Equation (4.2) on page. You can state
the equation for marginal probability by using the general multiplication rule. If
P1A2 = P1A and B12 + P1A and B22 + + P1A and Bk2
then, using the general multiplication rule, Equation (4.8) defines the marginal probability.
MARGINAL PROBABILITY USING THE GENERAL MULTIPLICATION RULE
P1A2 = P1A|B12P1B12 + P1A|B22P1B22 + + P1A|Bk2P1Bk2 (4.8)
where B1, B2, , Bk are k mutually exclusive and collectively exhaustive events.
To illustrate Equation (4.8), refer to Table 4.1 on page. Let

P1A2 = probability of planned to purchase
P1B12 = probability of actually purchased
P1B22 = probability of did not actually purchase
Then, using Equation (4.8), the probability of planned to purchase is
P1A2 = P1A|B12P1B12 + P1A|B22P1B22
= a ba b + a ba b
200 300 50 700
300 1,000 700 1,000
200 50 250
= + = = 0.25
1,000 1,000 1,000
Problems for Sections 4.2 161

LEARNING THE BASICS b. Given that the year 2009 is involved, what is the proba-
4.16 Consider the following contingency table: bility that three or more clicks are needed to be removed
from an email list?
c. Explain the difference in the results in (a) and (b).
B B d. Are needing three or more clicks to be removed from an
A 10 20 email list and the year independent?
A 20 40 4.22 Do people of different age groups differ in their re-
sponse to email messages? A survey by the Center for the
Digital Future of the University of Southern California (data
What is the probability of extracted from A. Mindlin, Older E-mail Users Favor Fast
a. A|B? Replies, The New York Times, July 14, 2008, p. B3) re-
b. A|B? ported that 70.7% of users over 70 years of age believe that
c. A|B? email messages should be answered quickly, as compared to
d. Are events A and B independent? 53.6% of users 12 to 50 years old. Suppose that the survey
4.17 Consider the following contingency table: was based on 1,000 users over 70 years of age and 1,000
users 12 to 50 years old. The following table summarizes the
results:
B B
A 10 30
A 25 35 ANSWERS QUICKLY
AGE OF RESPONDENTS 1250 Over 70 Total
What is the probability of Yes 536 707 1,243
a. A|B? No 464 293 757
b. A|B? Total 1,000 1,000 2,000
c. A|B?
d. Are events A and B independent?
4.18 If P1A and B2 = 0.4 and P1B2 = 0.8, find P1A|B2. a. Suppose you know that the respondent is between 12 and
50 years old. What is the probability that he or she an-
4.19 If P1A2 = 0.7, P1B2 = 0.6, and A and B are inde- swers quickly?
pendent, find P(A and B). b. Suppose you know that the respondent is over 70 years
4.20 If P1A2 = 0.3, P1B2 = 0.4, and P1A and B2 = 0.2, old. What is the probability that he or she answers
are A and B independent? quickly?
c. Are the two events, answers quickly and age of respon-
APPLYING THE CONCEPTS dents, independent? Explain.
4.21 Does it take more time to be removed from an email 4.23 What is the preferred way for people to order fast
list than it used to take? A study of 100 large online retailers food? A survey was conducted in 2009, but the sample sizes
revealed the following: were not reported. Suppose the results, based on a sample of
100 males and 100 females, were as follows:
NEED THREE OR MORE CLICKS

TO BE REMOVED GENDER
YEAR Yes No DINING PREFERENCE Male Female Total
2009 39 61 Dine inside 21 12 33
2008 7 93 Order inside to go 19 10 29
Source: Data extracted from More Clicks to Escape an Email List, Order at the
The New York Times, March 29, 2010, p. B2. drive-through 60 78 138
Total 100 100 200
a. Given that three or more clicks are needed to be removed Source: Data extracted from www.qsrmagazine.com/reports/drive-
thru_time_study/2009/2009_charts/whats_your_preferred_way_to
from an email list, what is the probability that this _order_fast_food.html.
occurred in 2009?
a. Given that a respondent is a male, what is the probability 4.26 Each year, ratings are compiled concerning the
that he prefers to order at the drive-through? performance of new cars during the first 90 days of use.
b. Given that a respondent is a female, what is the probabil- Suppose that the cars have been categorized according to
ity that she prefers to order at the drive-through? whether a car needs warranty-related repair (yes or no) and
c. Is dining preference independent of gender? Explain. the country in which the company manufacturing a car is
based (United States or not United States). Based on the
SELF 4.24 According to a Gallup Poll, the extent to
Test data collected, the probability that the new car needs a
which employees are engaged with their work-
warranty repair is 0.04, the probability that the car is manu-
place varies from country to country. Gallup reports that the
factured by a U.S.-based company is 0.60, and the probabil-
percentage of U.S. workers engaged with their workplace is
ity that the new car needs a warranty repair and was
more than twice as high as the percentage of German work-
manufactured by a U.S.-based company is 0.025.
ers. The study also shows that having more engaged work-
a. Suppose you know that a company based in the United
ers leads to increased innovation, productivity, and
States manufactured a particular car. What is the proba-
profitability, as well as reduced employee turnover. The
bility that the car needs warranty repair?
results of the poll are summarized in the following table:
b. Suppose you know that a company based in the United
States did not manufacture a particular car. What is the
COUNTRY probability that the car needs warranty repair?
ENGAGEMENT United States Germany Total c. Are need for warranty repair and location of the company
manufacturing the car independent?
Engaged 550 246 796
Not engaged 1,345 1,649 2,994 4.27 In 38 of the 60 years from 1950 through 2009, the
Total 1,895 1,895 3,790 S&P 500 finished higher after the first five days of trading.
In 33 of those 38 years, the S&P 500 finished higher for the
Source: Data extracted from M. Nink, Employee Disengagement year. Is a good first week a good omen for the upcoming
Plagues Germany, Gallup Management Journal, gmj.gallup.com, year? The following table gives the first-week and annual
April 9, 2009.
performance over this 60-year period:
a. Given that a worker is from the United States, what is the

probability that the worker is engaged? S&P 500S ANNUAL PERFORMANCE
b. Given that a worker is from the United States, what is the
FIRST WEEK Higher Lower
probability that the worker is not engaged?
c. Given that a worker is from Germany, what is the proba- Higher 33 5
bility that the worker is engaged? Lower 11 11
d. Given that a worker is from Germany, what is the proba-
bility that the worker is not engaged?
4.25 A sample of 500 respondents in a large metropolitan a. If a year is selected at random, what is the probability
area was selected to study consumer behavior, with the fol- that the S&P 500 finished higher for the year?
lowing results: b. Given that the S&P 500 finished higher after the first
five days of trading, what is the probability that it
finished higher for the year?
GENDER c. Are the two events first-week performance and an-
ENJOYS SHOPPING
FOR CLOTHING Male Female Total nual performance independent? Explain.
d. Look up the performance after the first five days of 2010
Yes 136 224 360 and the 2010 annual performance of the S&P 500 at
No 104 36 140 finance.yahoo.com. Comment on the results.
Total 240 260 500
4.28 A standard deck of cards is being used to play a
game. There are four suits (hearts, diamonds, clubs, and
a. Suppose that the respondent chosen is a female. What is spades), each having 13 faces (ace, 2, 3, 4, 5, 6, 7, 8, 9, 10,
the probability that she does not enjoy shopping for jack, queen, and king), making a total of 52 cards. This
clothing? complete deck is thoroughly mixed, and you will receive the
b. Suppose that the respondent chosen enjoys shopping for first 2 cards from the deck, without replacement (the first
clothing. What is the probability that the individual is a card is not returned to the deck after it is selected).
male? a. What is the probability that both cards are queens?
c. Are enjoying shopping for clothing and the gender of the b. What is the probability that the first card is a 10 and the
individual independent? Explain. second card is a 5 or 6?
4.3 Bayes Theorem 163
c. If you were sampling with replacement (the first card is after it is selected), what is the probability that both
returned to the deck after it is selected), what would be gloves selected will be right-handed?
the answer in (a)? b. If two gloves are randomly selected from the box, without
d. In the game of blackjack, the face cards (jack, queen, replacement (the first glove is not returned to the box after
king) count as 10 points, and the ace counts as either 1 or it is selected), what is the probability that there will be one
11 points. All other cards are counted at their face value. right-handed glove and one left-handed glove selected?
Blackjack is achieved if 2 cards total 21 points. What is c. If three gloves are selected, with replacement (the gloves
the probability of getting blackjack in this problem? are returned to the box after they are selected), what is
the probability that all three will be left-handed?
4.29 A box of nine gloves contains two left-handed gloves
d. If you were sampling with replacement (the first glove is
and seven right-handed gloves.
returned to the box after it is selected), what would be the
a. If two gloves are randomly selected from the box, with-
answers to (a) and (b)?
out replacement (the first glove is not returned to the box
4.3 Bayes Theorem

Bayes theorem is used to revise previously calculated probabilities based on new informa-
tion. Developed by Thomas Bayes in the eighteenth century (see references 1, 2, and 7), Bayes
theorem is an extension of what you previously learned about conditional probability.
You can apply Bayes theorem to the situation in which M&R Electronics World is consid-
ering marketing a new model of televisions. In the past, 40% of the new-model televisions
have been successful, and 60% have been unsuccessful. Before introducing the new model tel-
evision, the marketing research department conducts an extensive study and releases a report,
either favorable or unfavorable. In the past, 80% of the successful new-model television(s) had
received favorable market research reports, and 30% of the unsuccessful new-model televi-
sion(s) had received favorable reports. For the new model of television under consideration,
the marketing research department has issued a favorable report. What is the probability that
the television will be successful?
Bayes theorem is developed from the definition of conditional probability. To find the con-
ditional probability of B given A, consider Equation (4.4b) (originally presented on page 155
and shown below):
P1A and B2 P1A|B2P1B2
P1B|A2 = =
P1A2 P1A2
Bayes theorem is derived by substituting Equation (4.8) on page 160 for P(A) in the denomi-
nator of Equation (4.4b).
BAYES THEOREM
P1A|Bi2P1Bi2
P1Bi|A2 =
P1A|B12P1B12 + P1A|B22P1B22 + + P1A|Bk2P1Bk2
(4.9)
where Bi is the ith event out of k mutually exclusive and collectively exhaustive events.
To use Equation (4.9) for the television-marketing example, let

event S = successful television event F = favorable report
event S = unsuccessful television event F = unfavorable report
and
P1S2 = 0.40 P1F|S2 = 0.80
P1S2 = 0.60 P1F|S2 = 0.30
Then, using Equation (4.9),

P1F|S2P1S2
P1S|F2 =
P1F|S2P1S2 + P1F|S2P1S2
10.80210.402
=
10.80210.402 + 10.30210.602
0.32 0.32
= =
0.32 + 0.18 0.50
= 0.64
The probability of a successful television, given that a favorable report was received, is
0.64. Thus, the probability of an unsuccessful television, given that a favorable report was
received, is 1 - 0.64 = 0.36.
Table 4.4 summarizes the computation of the probabilities, and Figure 4.5 presents the
decision tree.
TA B L E 4 . 4 Prior Conditional Joint

Bayes Theorem Probability Probability Probability Revised Probability
Calculations for the Event Si P(Si ) P(F Si ) P(F Si )P(Si ) P(Si F)
Television-Marketing
Example
S successful 0.40 0.80 0.32 P1S|F2 = 0.32>0.50
television = 0.64
S unsuccessful 0.60 0.30 0.18 P1S|F2 = 0.18>0.50
television 0.50 = 0.36
FIGURE 4.5
P(S and F) = P(F|S) P(S)
Decision tree for = (0.80) (0.40) = 0.32
marketing a new
television P(S) = 0.40

= (0.20) (0.40) = 0.08

= (0.30) (0.60) = 0.18
P(S) = 0.60

= (0.70) (0.60) = 0.42
Example 4.10 applies Bayes theorem to a medical diagnosis problem.
EXAMPLE 4.10 The probability that a person has a certain disease is 0.03. Medical diagnostic tests are avail-
able to determine whether the person actually has the disease. If the disease is actually
Using Bayes present, the probability that the medical diagnostic test will give a positive result (indicating
Theorem in a that the disease is present) is 0.90. If the disease is not actually present, the probability of a
Medical Diagnosis positive test result (indicating that the disease is present) is 0.02. Suppose that the medical
Problem diagnostic test has given a positive result (indicating that the disease is present). What is the
probability that the disease is actually present? What is the probability of a positive test
result?
4.3 Bayes Theorem 165
SOLUTION Let
event D = has disease event T = test is positive
event D = does not have disease event T = test is negative
and
P1D2 = 0.03 P1T|D2 = 0.90
P1D2 = 0.97 P1T|D2 = 0.02
Using Equation (4.9) on page 163,
P1T|D2P1D2
P1D|T2 =
P1T|D2P1D2 + P1T|D2P1D2
10.90210.032
=
10.90210.032 + 10.02210.972
0.0270 0.0270
= =
0.0270 + 0.0194 0.0464
= 0.582
The probability that the disease is actually present, given that a positive result has occurred
(indicating that the disease is present), is 0.582. Table 4.5 summarizes the computation of the
probabilities, and Figure 4.6 presents the decision tree.
TA B L E 4 . 5 Prior Conditional Joint

Bayes Theorem Probability Probability Probability Revised Probability
Calculations for the Event Di P(Di) P(T | Di) P(T | Di)P(Di) P(Di | T)
Medical Diagnosis
Problem
D has 0.03 0.90 0.0270 P(D|T2 = 0.0270>0.0464
disease = 0.582
D does not 0.97 0.02 0.0194 P(D|T 2 = 0.0194>0.0464
have disease 0.0464 = 0.418
FIGURE 4.6
P(D and T ) = P(T |D) P(D)
Decision tree for the = (0.90) (0.03) = 0.0270
medical diagnosis
problem P(D) = 0.03

= (0.10) (0.03) = 0.0030

= (0.02) (0.97) = 0.0194
P(D) = 0.97

= (0.98) (0.97) = 0.9506
The denominator in Bayes theorem represents P(T ), the probability of a positive test re-
sult, which in this case is 0.0464, or 4.64%.
THINK ABOUT THIS Divine Providence and Spam

Would you ever guess that the essays Divine We can use a similar process to try to clas- overlooked the fact that the conditional probabil-
Benevolence: Or, An Attempt to Prove That the sify a new email message in your in-box as ities are constantly updated and that words once
Principal End of the Divine Providence and Gov- either spam or a legitimate message (called considered good would be soon discarded
ernment Is the Happiness of His Creatures and ham, in this context). We would first need to from the good list by the filter as their P(A|B)
An Essay Towards Solving a Problem in the Doc- add to your email program a spam filter that value increased. Likewise, as rare words grew
trine of Chances were written by the same per- has the ability to track word frequencies associ- more common in spam and yet stayed rare in
son? Probably not, and in doing so, you illustrate ated with spam and ham messages as you iden- ham, such words acted like the misspelled vari-
a modern-day application of Bayesian statistics: tify them on a day-to-day basis. This would allow ants that others had tried earlier.
spam, or junk mail, filters. the filter to constantly update the prior probabili- Even then, and perhaps after reading about
In not guessing correctly, you probably ties necessary to use Bayes theorem. With these Bayesian statistics, spammers thought that
looked at the words in the titles of the essays probabilities, the filter can ask, What is the they could break Bayesian filters by inserting
and concluded that they were talking about two probability that an email is spam, given the pres- random words in their messages. Those ran-
different things. An implicit rule you used was ence of a certain word? dom words would affect the filter by causing it
that word frequencies vary by subject matter. A Applying the terms of Equation (4.9) to see many words whose P(A|B) value would
statistics essay would very likely contain the on page 163, such a Bayesian spam filter would be low. The Bayesian filter would begin to label
word statistics as well as words such as chance, multiply the probability of finding the word in a many spam messages as ham and end up
problem, and solving. An eighteenth-century es- spam email, P(A |B), by the probability that the being of no practical use. Spammers again
say about theology and religion would be more email is spam, P(B), and then divide by the prob- overlooked that conditional probabilities are
likely to contain the uppercase forms of Divine ability of finding the word in an email, the de- constantly updated.
and Providence. nominator in Equation (4.9). Bayesian spam Other spammers decided to eliminate all or
Likewise, there are words you would guess filters also use shortcuts by focusing on a small most of the words in their messages and replace
to be very unlikely to appear in either book, such set of words that have a high probability of being them with graphics so that Bayesian filters
as technical terms from finance, and words that found in a spam message as well as on a small would have very few words with which to form
are most likely to appear in bothcommon set of other words that have a low probability of conditional probabilities. But this approach
words such as a, and, and the. That words would being found in a spam message. failed, too, as Bayesian filters were rewritten to
either be likely or unlikely suggests an applica- As spammers (people who send junk email) consider things other than words in a message.
tion of probability theory. Of course, likely and learned of such new filters, they tried to outfox After all, Bayes theorem concerns events, and
unlikely are fuzzy concepts, and we might occa- them. Having learned that Bayesian filters might graphics present with no text is as valid an
sionally misclassify an essay if we kept things be assigning a high P(A|B) value to words com- event as some word, X, present in a message.
too simple, such as relying solely on the occur- monly found in spam, such as Viagra, spammers Other future tricks will ultimately fail for the
rence of the words Divine and Providence. thought they could fool the filter by misspelling same reason. (By the way, spam filters use non-
For example, a profile of the late Harris Mil- the word as Vi@gr@ or V1agra. What they over- Bayesian techniques as well, which makes
stead, better known as Divine, the star of looked was that the misspelled variants were spammers lives even more difficult.)
Hairspray and other films, visiting Providence even more likely to be found in a spam message Bayesian spam filters are an example of the
(Rhode Island), would most certainly not be an than the original word. Thus, the misspelled vari- unexpected way that applications of statistics
essay about theology. But if we widened the ants made the job of spotting spam easier for the can show up in your daily life. You will discover
number of words we examined and found such Bayesian filters. more examples as you read the rest of this book.
words as movie or the name John Waters (Di- Other spammers tried to fool the filters by By the way, the author of the two essays men-
vines director in many films), we probably would adding good words, words that would have a tioned earlier was Thomas Bayes, who is a lot
quickly realize the essay had something to do low probability of being found in a spam mes- more famous for the second essay than the first
with twentieth-century cinema and little to do sage, or rare words, words not frequently en- essay, a failed attempt to use mathematics and
with theology and religion. countered in any message. But these spammers logic to prove the existence of God.

LEARNING THE BASICS tive result if the disease is not present is reduced from
4.30 If P1B2 = 0.05, P1A|B2 = 0.80, P1B2 = 0.95, and 0.02 to 0.01.
P1A|B2 = 0.40, find P1B|A2. a. If the medical diagnostic test has given a positive result
(indicating that the disease is present), what is the proba-
4.31 If P1B2 = 0.30, P1A|B2 = 0.60, P1B2 = 0.70, and bility that the disease is actually present?
P1A|B2 = 0.50, find P1B|A2. b. If the medical diagnostic test has given a negative result
(indicating that the disease is not present), what is the
APPLYING THE CONCEPTS probability that the disease is not present?
4.32 In Example 4.10 on page 164, suppose that the 4.33 An advertising executive is studying television viewing
probability that a medical diagnostic test will give a posi- habits of married men and women during prime-time hours.
4.4 Counting Rules 167
Based on past viewing records, the executive has determined a. If an entrepreneur is selected at random and that individ-
that during prime time, husbands are watching television 60% ual expects that his or her new business will employ 20
of the time. When the husband is watching television, 40% of or more people within five years, what is the probability
the time the wife is also watching. When the husband is not that this individual is an entrepreneur by choice?
watching television, 30% of the time the wife is watching tel- b. Discuss several possible reasons why entrepreneurs by
evision. choice are more likely than entrepreneurs by necessity to
a. Find the probability that if the wife is watching televi- believe that they will grow their businesses.
sion, the husband is also watching television.
4.36 The editor of a textbook publishing company is try-
b. Find the probability that the wife is watching television
ing to decide whether to publish a proposed business statis-
during prime time.
tics textbook. Information on previous textbooks published
SELF 4.34 Olive Construction Company is determin- indicates that 10% are huge successes, 20% are modest
Test ing whether it should submit a bid for a new successes, 40% break even, and 30% are losers. However,
shopping center. In the past, Olives main competitor, Base before a publishing decision is made, the book will be
Construction Company, has submitted bids 70% of the time. reviewed. In the past, 99% of the huge successes received
If Base Construction Company does not bid on a job, the favorable reviews, 70% of the moderate successes received
probability that Olive Construction Company will get favorable reviews, 40% of the break-even books received fa-
the job is 0.50. If Base Construction Company bids on a job, vorable reviews, and 20% of the losers received favorable
the probability that Olive Construction Company will reviews.
get the job is 0.25. a. If the proposed textbook receives a favorable review, how
a. If Olive Construction Company gets the job, what is the should the editor revise the probabilities of the various
probability that Base Construction Company did not outcomes to take this information into account?
bid? b. What proportion of textbooks receives favorable
b. What is the probability that Olive Construction Company reviews?
will get the job?
4.37 A municipal bond service has three rating cate-
4.35 Laid-off workers who become entrepreneurs because gories (A, B, and C). Suppose that in the past year, of the
they cannot find meaningful employment with another com- municipal bonds issued throughout the United States,
pany are known as entrepreneurs by necessity. The Wall 70% were rated A, 20% were rated B, and 10% were rated
Street Journal reports that these entrepreneurs by necessity C. Of the municipal bonds rated A, 50% were issued by
are less likely to grow into large businesses than are cities, 40% by suburbs, and 10% by rural areas. Of the
entrepreneurs by choice (J. Bailey, DesireMore Than municipal bonds rated B, 60% were issued by cities, 20%
NeedBuilds a Business, The Wall Street Journal, May by suburbs, and 20% by rural areas. Of the municipal
21, 2001, p. B4). This article states that 89% of the entrepre- bonds rated C, 90% were issued by cities, 5% by suburbs,
neurs in the United States are entrepreneurs by choice and and 5% by rural areas.
11% are entrepreneurs by necessity. Only 2% of entrepre- a. If a new municipal bond is to be issued by a city, what is
neurs by necessity expect their new business to employ 20 the probability that it will receive an A rating?
or more people within five years, whereas 14% of entrepre- b. What proportion of municipal bonds are issued by cities?
neurs by choice expect to employ at least 20 people within c. What proportion of municipal bonds are issued by
five years. suburbs?
4.4 Counting Rules

In Equation (4.1) on page 146, the probability of occurrence of an outcome was defined as the
number of ways the outcome occurs, divided by the total number of possible outcomes. Often,
there are a large number of possible outcomes, and determining the exact number can be diffi-
cult. In such circumstances, rules have been developed for counting the number of possible
outcomes. This section presents five different counting rules.
Counting Rule 1
Counting rule 1 determines the number of possible outcomes for a set of mutually exclusive
and collectively exhaustive events.
COUNTING RULE 1
If any one of k different mutually exclusive and collectively exhaustive events can occur
on each of n trials, the number of possible outcomes is equal to
kn (4.10)
For example, using Equation (4.10), the number of different possible outcomes from toss-
ing a two-sided coin five times is 25 = 2 * 2 * 2 * 2 * 2 = 32.
EXAMPLE 4.11 Suppose you roll a die twice. How many different possible outcomes can occur?
Rolling a Die Twice SOLUTION If a six-sided die is rolled twice, using Equation (4.10), the number of different
outcomes is 62 = 36.
Counting Rule 2
The second counting rule is a more general version of the first and allows the number of possi-
ble events to differ from trial to trial.
COUNTING RULE 2
If there are k1 events on the first trial, k2 events on the second trial, ... , and kn events on
the nth trial, then the number of possible outcomes is
1k121k22 1kn2 (4.11)
For example, a state motor vehicle department would like to know how many license
plate numbers are available if a license plate number consists of three letters followed by
three numbers (0 through 9). Using Equation (4.11), if a license plate number consists of
three letters followed by three numbers, the total number of possible outcomes is
126212621262110211021102 = 17,576,000.
EXAMPLE 4.12 A restaurant menu has a price-fixed complete dinner that consists of an appetizer, an entre, a
beverage, and a dessert. You have a choice of 5 appetizers, 10 entres, 3 beverages, and 6
Determining the desserts. Determine the total number of possible dinners.
Number of Different
Dinners SOLUTION Using Equation (4.11), the total number of possible dinners is
1521102132162 = 900.
Counting Rule 3
The third counting rule involves computing the number of ways that a set of items can be
arranged in order.
4.4 Counting Rules 169
COUNTING RULE 3
The number of ways that all n items can be arranged in order is
n! = 1n21n - 12 112 (4.12)
where n! is called n factorial, and 0! is defined as 1.
EXAMPLE 4.13 If a set of six books is to be placed on a shelf, in how many ways can the six books be
arranged?
Using Counting
Rule 3 SOLUTION To begin, you must realize that any of the six books could occupy the first po-
sition on the shelf. Once the first position is filled, there are five books to choose from in
filling the second position. You continue this assignment procedure until all the positions
are occupied. The number of ways that you can arrange six books is
n! = 6! = 162152142132122112 = 720
Counting Rule 4
In many instances you need to know the number of ways in which a subset of an entire
group of items can be arranged in order. Each possible arrangement is called a
permutation.
COUNTING RULE 4: PERMUTATIONS

The number of ways of arranging x objects selected from n objects in order is
n!
=
1n - x2!
nPx (4.13)
where
n = total number of objects
x = number of objects to be arranged
1
On many scientific calculators,
there is a button labeled nPr that al- n! = n factorial = n1n - 12 112
lows you to compute permutations.
The symbol r is used instead of x. P = symbol for permutations1
EXAMPLE 4.14 Modifying Example 4.13, if you have six books, but there is room for only four books on the
shelf, in how many ways can you arrange these books on the shelf?
Using Counting
Rule 4 SOLUTION Using Equation (4.13), the number of ordered arrangements of four books
selected from six books is equal to
n! 6! 162152142132122112
= = = = 360
1n - x2! 16 - 42! 122112
nPx
Counting Rule 5
In many situations, you are not interested in the order of the outcomes but only in the number
of ways that x items can be selected from n items, irrespective of order. Each possible selec-
tion is called a combination.
COUNTING RULE 5: COMBINATIONS

The number of ways of selecting x objects from n objects, irrespective of order, is equal to
n!
nCx = (4.14)
x!1n - x2!
where
n = total number of objects
x = number of objects to be arranged
2
On many scientific calculators,
there is a button labeled nCr that al- n! = n factorial = n1n - 12 112
lows you to compute permutations.
The symbol r is used instead of x. C = symbol for combinations2
If you compare this rule to counting rule 4, you see that it differs only in the inclusion of a
term x! in the denominator. When permutations were used, all of the arrangements of the x
objects are distinguishable. With combinations, the x! possible arrangements of objects are
irrelevant.
EXAMPLE 4.15 Modifying Example 4.14, if the order of the books on the shelf is irrelevant, in how many ways
can you arrange these books on the shelf?
Using Counting
SOLUTION Using Equation (4.14), the number of combinations of four books selected from
Rule 5
six books is equal to
n! 6! 162152142132122112
= = = = 15
142132122112122112
nCx
x!1n - x2! 4!16 - 42!

APPLYING THE CONCEPTS 4.41 A particular brand of womens jeans is available in
4.38 If there are 10 multiple-choice questions seven different sizes, three different colors, and three differ-
SELF
Test on an exam, each having three possible answers, ent styles. How many different womens jeans does the store
how many different sequences of answers are there? manager need to order to have one pair of each type?
4.39 A lock on a bank vault consists of three dials, each 4.42 You would like to make a salad that consists of
with 30 positions. In order for the vault to open, each of the lettuce, tomato, cucumber, and peppers. You go to the super-
three dials must be in the correct position. market, intending to purchase one variety of each of these
a. How many different possible dial combinations are there ingredients. You discover that there are eight varieties of let-
for this lock? tuce, four varieties of tomatoes, three varieties of cucum-
b. What is the probability that if you randomly select a posi- bers, and three varieties of peppers for sale at the
tion on each dial, you will be able to open the bank vault? supermarket. If you buy them all, how many different salads
c. Explain why dial combinations are not mathematical can you make?
combinations expressed by Equation (4.14). 4.43 A team is being formed that includes four different
people. There are four different positions on the teams. How
4.40 a. If a coin is tossed seven times, how many different
many different ways are there to assign the four people to
outcomes are possible?
the four positions??
b. If a die is tossed seven times, how many different out-
comes are possible? 4.44 In Major League Baseball, there are five teams in the
c. Discuss the differences in your answers to (a) and (b). Eastern Division of the National League: Atlanta, Florida,
4.5 Ethical Issues and Probability 171
New York, Philadelphia, and Washington. How many differ- 4.48 Four members of a group of 10 people are to be se-
ent orders of finish are there for these five teams? (Assume lected to a team. How many ways are there to select these
that there are no ties in the standings.) Do you believe that four members?
all these orders are equally likely? Discuss.
4.49 A student has seven books that she would like to
4.45 Referring to Problem 4.44, how many different orders place in her backpack. However, there is room for only four
of finish are possible for the first four positions? books. Regardless of the arrangement, how many ways are
there of placing four books into the backpack?
4.46 A gardener has six rows available in his vegetable
garden to place tomatoes, eggplant, peppers, cucumbers, 4.50 A daily lottery is conducted in which 2 winning num-
beans, and lettuce. Each vegetable will be allowed one and bers are selected out of 100 numbers. How many different
only one row. How many ways are there to position these combinations of winning numbers are possible?
vegetables in this garden?
4.51 A reading list for a course contains 20 articles.
4.47 There are eight members of a team. How many ways How many ways are there to choose 3 articles from this
are there to select a team leader, assistant team leader, and list?
team coordinator?
4.5 Ethical Issues and Probability

Ethical issues can arise when any statements related to probability are presented to the public,
particularly when these statements are part of an advertising campaign for a product or service.
Unfortunately, many people are not comfortable with numerical concepts (see reference 5) and
tend to misinterpret the meaning of the probability. In some instances, the misinterpretation is
not intentional, but in other cases, advertisements may unethically try to mislead potential
customers.
One example of a potentially unethical application of probability relates to advertise-
ments for state lotteries. When purchasing a lottery ticket, the customer selects a set of num-
bers (such as 6) from a larger list of numbers (such as 54). Although virtually all participants
know that they are unlikely to win the lottery, they also have very little idea of how unlikely it
is for them to select all 6 winning numbers from the list of 54 numbers. They have even less
of an idea of the probability of winning a consolation prize by selecting either 4 or 5 winning
numbers.
Given this background, you might consider a recent commercial for a state lottery that
stated, We wont stop until we have made everyone a millionaire to be deceptive and possi-
bly unethical. Do you think the state has any intention of ever stopping the lottery, given the
fact that the state relies on it to bring millions of dollars into its treasury? Is it possible that the
lottery can make everyone a millionaire? Is it ethical to suggest that the purpose of the lottery
is to make everyone a millionaire?
Another example of a potentially unethical application of probability relates to an
investment newsletter promising a 90% probability of a 20% annual return on investment.
To make the claim in the newsletter an ethical one, the investment service needs to (a) ex-
plain the basis on which this probability estimate rests, (b) provide the probability
statement in another format, such as 9 chances in 10, and (c) explain what happens to the
investment in the 10% of the cases in which a 20% return is not achieved (e.g., is the entire
investment lost?).
These are serious ethical issues. If you were going to write an advertisement for the state
lottery that ethically describes the probability of winning a certain prize, what would you say?
If you were going to write an advertisement for the investment newsletter that ethically states
the probability of a 20% return on an investment, what would you say?
U S I N G S TAT I S T I C S @ M&R Electronics World Revisited
A s the marketing manager for M&R Electronics World, you

analyzed the survey results of an intent-to-purchase study. This
study asked the heads of 1,000 households about their intentions
to purchase a big-screen television sometime during the next
12 months, and as a follow-up, M&R surveyed the same people 12 months later
to see whether such a television was purchased. In addition, for households
purchasing big-screen televisions, the survey asked whether the television they
purchased had a faster refresh rate, whether they also purchased a Blu-ray disc (BD) player in the
past 12 months, and whether they were satisfied with their purchase of the big-screen television.
By analyzing the results of these surveys, you were able to uncover many pieces of valu-
able information that will help you plan a marketing strategy to enhance sales and better target
those households likely to purchase multiple or more expensive products. Whereas only 30%
of the households actually purchased a big-screen television, if a household indicated that it
planned to purchase a big-screen television in the next 12 months, there was an 80% chance
that the household actually made the purchase. Thus the marketing strategy should target those
households that have indicated an intention to purchase.
You determined that for households that purchased a television that had a faster refresh
rate, there was a 47.5% chance that the household also purchased a Blu-ray disc player. You
then compared this conditional probability to the marginal probability of purchasing a Blu-ray
disc player, which was 36%. Thus, households that purchased televisions that had a faster
refresh rate are more likely to purchase a Blu-ray disc player than are households that pur-
chased big-screen televisions that have a standard refresh rate.
You were also able to apply Bayes theorem to M&R Electronics Worlds market research
reports. The reports investigate a potential new television model prior to its scheduled release.
If a favorable report was received, then there was a 64% chance that the new television model
would be successful. However, if an unfavorable report was received, there is only a 16%
chance that the model would be successful. Therefore, the marketing strategy of M&R needs
to pay close attention to whether a reports conclusion is favorable or unfavorable.
SUMMARY
This chapter began by developing the basic concepts of previously calculated probabilities based on new information.
probability. You learned that probability is a numeric value You also learned about several counting rules. Throughout
from 0 to 1 that represents the chance, likelihood, or possi- the chapter, contingency tables and decision trees were used
bility that a particular event will occur. In addition to simple to display information. In the next chapter, important dis-
probability, you learned about conditional probabilities and crete probability distributions such as the binomial, Poisson,
independent events. Bayes theorem was used to revise and hypergeometric distributions are developed.
K E Y E Q U AT I O N S
Probability of Occurrence Conditional Probability
X P1A and B2
Probability of occurrence = (4.1) P1A|B2 = (4.4a)
T P1B2
Marginal Probability P1A and B2
P1B|A2 =
P1A2 = P1A and B12 + P1A and B22
(4.4b)
P1A2
+ + P1A and Bk2 (4.2)
General Addition Rule Independence
P1A or B2 = P1A2 + P1B2 - P1A and B2 (4.3) P1A|B2 = P1A2 (4.5)
Chapter Review Problems 173
General Multiplication Rule Counting Rule 1

P1A and B2 = P1A|B2P1B2 (4.6) kn (4.10)
Multiplication Rule for Independent Events Counting Rule 2
P1A and B2 = P1A2P1B2 (4.7) 1k121k22 1kn2 (4.11)
Marginal Probability Using the General Multiplication Rule Counting Rule 3
P1A2 = P1A|B12P1B12 + P1A|B22P1B22 n! = 1n21n - 12 112 (4.12)

+ + P1A|Bk2P1Bk2 (4.8)
Counting Rule 4: Permutations
Bayes Theorem n!
=
1n - x2!
nPx (4.13)
P1Bi|A2 =
P1A|Bi2P1Bi2 Counting Rule 5: Combinations
P1A|B12P1B12 + P1A|B22P1B22 + + P1A|Bk2P1Bk2 =
n!
nCx (4.14)
(4.9) x!1n - x2!
KEY TERMS
a priori probability 146 event 147 mutually exclusive 151
Bayes theorem 163 general addition rule 152 permutation 169
certain event 146 general multiplication rule 159 probability 146
collectively exhaustive 151 impossible event 146 sample space 148
combination 169 independence 158 simple event 147
complement 147 joint event 147 simple probability 149
conditional probability 155 joint probability 150 subjective probability 147
contingency table 148 marginal probability 150 Venn diagram 148
decision tree 156 multiplication rule for independent
empirical probability 147 events 160
CHAPTER REVIEW PROBLEMS

CHECKING YOUR UNDERSTANDING APPLYING THE CONCEPTS
4.52 What are the differences between a priori probabil- 4.60 A survey by the Pew Research Center (Snapshots:
ity, empirical probability, and subjective probability? Goals of Gen Next vs. Gen X, USA Today, March 27,
2007, p. 1A) indicated that 81% of 18- to 25-year-olds had
4.53 What is the difference between a simple event and a
getting rich as a goal, as compared to 62% of 26- to 40-year-
joint event?
olds. Suppose that the survey was based on 500 respondents
4.54 How can you use the general addition rule to find the from each of the two groups.
probability of occurrence of event A or B? a. Construct a contingency table.
4.55 What is the difference between mutually exclusive b. Give an example of a simple event and a joint event.
events and collectively exhaustive events? c. What is the probability that a randomly selected respon-
dent has a goal of getting rich?
4.56 How does conditional probability relate to the con- d. What is the probability that a randomly selected
cept of independence? respondent has a goal of getting rich and is in the
4.57 How does the multiplication rule differ for events that 26- to 40-year-old group?
are and are not independent? e. Are the events age group and has getting rich as a
goal independent? Explain.
4.58 How can you use Bayes theorem to revise probabili-
ties in light of new information? 4.61 The owner of a restaurant serving Continental-style
entres was interested in studying ordering patterns of pa-
4.59 In Bayes theorem, how does the prior probability
trons for the Friday-to-Sunday weekend time period.
differ from the revised probability?
Records were maintained that indicated the demand for d. is a male and prefers ordering dinner at the drive-
dessert during the same time period. The owner decided to through?
study two other variables, along with whether a dessert was e. Given that the person selected is a female, what is the
ordered: the gender of the individual and whether a beef probability that she prefers ordering breakfast at the
entre was ordered. The results are as follows: drive-through?
4.63 According to a Gallup Poll, companies with
GENDER
employees who are engaged with their workplace have
DESSERT ORDERED Male Female Total greater innovation, productivity, and profitability, as well
as less employee turnover. A survey of 1,895 workers in
Yes 96 40 136
Germany found that 13% of the workers were engaged,
No 224 240 464
67% were not engaged, and 20% were actively disen-
Total 320 280 600
gaged. The survey also noted that 48% of engaged work-
ers strongly agreed with the statement My current job
BEEF ENTRE brings out my most creative ideas. Only 20% of the not
engaged workers and 3% of the actively disengaged work-
DESSERT ORDERED Yes No Total
ers agreed with this statement (data extracted from
Yes 71 65 136 M. Nink, Employee Disengagement Plagues Germany,
No 116 348 464 Gallup Management Journal, gmj.gallup.com, April 9,
Total 187 413 600 2009). If a worker is known to strongly agree with the
statement My current job brings out my most creative
A waiter approaches a table to take an order for dessert. ideas, what is the probability that the worker is engaged?
What is the probability that the first customer to order at the
4.64 Sport utility vehicles (SUVs), vans, and pickups are
table
generally considered to be more prone to roll over than cars.
a. orders a dessert?
In 1997, 24.0% of all highway fatalities involved rollovers;
b. orders a dessert or has ordered a beef entre?
15.8% of all fatalities in 1997 involved SUVs, vans, and
c. is a female and does not order a dessert?
pickups, given that the fatality involved a rollover. Given
d. is a female or does not order a dessert?
that a rollover was not involved, 5.6% of all fatalities in-
e. Suppose the first person from whom the waiter takes the
volved SUVs, vans, and pickups (data extracted from
dessert order is a female. What is the probability that she
A. Wilde Mathews, Ford Ranger, Chevy Tracker Tilt in
does not order dessert?
Test, The Wall Street Journal, July 14, 1999, p. A2).
f. Are gender and ordering dessert independent?
Consider the following definitions:
g. Is ordering a beef entre independent of whether the
person orders dessert? A = fatality involved an SUV, van, or pickup
4.62 Which meal are people most likely to order at a drive- B = fatality involved a rollover
through? A survey was conducted in 2009, but the sample
sizes were not reported. Suppose the results, based on a a. Use Bayes theorem to find the probability that a fatality
sample of 100 males and 100 females, were as follows: involved a rollover, given that the fatality involved an
SUV, a van, or a pickup.
GENDER b. Compare the result in (a) to the probability that a fatality
involved a rollover and comment on whether SUVs, vans,
MEAL Male Female Total and pickups are generally more prone to rollover acci-
Breakfast 18 10 28 dents than other vehicles.
Lunch 47 52 99 4.65 Enzyme-linked immunosorbent assay (ELISA) is the
Dinner 29 29 58 most common type of screening test for detecting the HIV
Snack/beverage 6 9 15 virus. A positive result from an ELISA indicates that the
Total 100 100 200 HIV virus is present. For most populations, ELISA has a
Source: Data extracted from www.qsrmagazine.com/reports/drive- high degree of sensitivity (to detect infection) and speci-
thru_time_study/2009/2009_charts/whats_your_preferred_way ficity (to detect noninfection). (See HIV InSite Gateway to
_to_order_fast_food.html. HIV and AIDS Knowledge at HIVInsite.ucsf.edu.) Sup-
pose the probability that a person is infected with the HIV
If a respondent is selected at random, what is the probability virus for a certain population is 0.015. If the HIV virus is
that he or she actually present, the probability that the ELISA test will
a. prefers ordering lunch at the drive-through? give a positive result is 0.995. If the HIV virus is not actu-
b. prefers ordering breakfast or lunch at the drive-through? ally present, the probability of a positive result from an
c. is a male or prefers ordering dinner at the drive-through? ELISA is 0.01. If the ELISA has given a positive result, use
Digital Case 175
Bayes theorem to find the probability that the HIV virus is a. Select a sample of undergraduate students at your school
actually present. and conduct a similar survey for those students.
b. For your data, construct contingency tables of gender
TEAM PROJECT and major, gender and graduate school intention, gender
and employment status, gender and computer prefer-
The file Bond Funds contains information regarding three ence, class and graduate school intention, class and em-
categorical variables from a sample of 184 bond funds. The ployment status, major and graduate school intention,
variables include major and employment status, and major and computer
TypeBond fund type (intermediate government or preference.
short-term corporate) c. Based on (b), what conclusions can you reach about
FeesSales charges (no or yes) whether these variables are independent?
RiskRisk-of-loss factor of the bond fund (below aver- d. Compare the results of (c) to those of Problem 4.67 (b).
age, average, or above average) 4.69 Problem 1.28 on page 15 describes a survey of
4.66 Construct contingency tables of type and fees, type 44 MBA students (stored in GradSurvey ). For these data,
and risk, and fees and risk. construct contingency tables of gender and graduate ma-
a. For each of these contingency tables, compute all the jor, gender and undergraduate major, gender and employ-
conditional and marginal probabilities. ment status, gender and computer preference, graduate
b. Based on (a), what conclusions can you reach about major and undergraduate major, graduate major and em-
whether these variables are independent? ployment status, and graduate major and computer
preference.
a. For each of these contingency tables, compute all the
STUDENT SURVEY DATABASE
conditional and marginal probabilities.
4.67 Problem 1.27 on page 14 describes a survey of 62 b. Based on (b), what conclusions can you reach about
undergraduate students (see the file UndergradSurvey ). whether these variables are independent?
For these data, construct contingency tables of gender and 4.70 Problem 1.28 on page 15 describes a survey of 44
major, gender and graduate school intention, gender and MBA students (stored in GradSurvey ).
employment status, gender and computer preference, a. Select a sample of MBA students from your MBA pro-
class and graduate school intention, class and employ- gram and conduct a similar survey for those students.
ment status, major and graduate school intention, major b. For your data, construct contingency tables of gender and
and employment status, and major and computer graduate major, gender and undergraduate major, gender
preference. and employment status, gender and computer preference,
a. For each of these contingency tables, compute all the graduate major and undergraduate major, graduate major
conditional and marginal probabilities. and employment status, and graduate major and com-
b. Based on (a), what conclusions can you reach about puter preference.
whether these variables are independent? c. Based on (b), what conclusions can you reach about
4.68 Problem 1.27 on page 14 describes a survey of 62 whether these variables are independent?
undergraduate students (stored in UndergradSurvey ). d. Compare the results of (c) to those of Problem 4.69 (b).
D I G I TA L C A S E
Apply your knowledge about contingency tables and the misleading? How would you calculate and state the
proper application of simple and joint probabilities in this probability of having an annual rate of return not less
continuing Digital Case from Chapter 3. than 15%?
2. Using the table found under the Show Me The Winning
Open EndRunGuide.pdf, the EndRun Financial Services
Probabilities subhead, compute the proper probabilities
Guide to Investing, and read the information about the
for the group of investors. What mistake was made in re-
Guaranteed Investment Package (GIP). Read the claims and
porting the 7% probability claim?
examine the supporting data. Then answer the following
questions: 3. Are there any probability calculations that would be
appropriate for rating an investment service? Why or why
1. How accurate is the claim of the probability of suc-
not?
cess for EndRuns GIP? In what ways is the claim
REFERENCES
1. Bellhouse, D. R., The Reverend Thomas Bayes, FRS: A 4. Minitab Release 16 (State College, PA.: Minitab, Inc.,
Biography to Celebrate the Tercentenary of His Birth, 2010).
Statistical Science, 19 (2004), 343. 5. Paulos, J. A., Innumeracy (New York: Hill and Wang,
2. Lowd, D., and C. Meek, Good Word Attacks on Statis- 1988).
tical Spam Filters, presented at the Second Conference 6. Silberman, S., The Quest for Meaning, Wired 8.02,
on Email and Anti-Spam, CEAS 2005. February 2000.
3. Microsoft Excel 2010 (Redmond, WA: Microsoft 7. Zeller, T., The Fight Against V1@gra (and Other
Corp., 2010). Spam), The New York Times, May 21, 2006, pp. B1, B6.
CHAPTER 4 EXCEL GUIDE
EG4.1 BASIC PROBABILITY CONCEPTS In-Depth Excel Use the COMPUTE worksheet of the
Probabilities workbook as a template for computing
Simple and Joint Probability and the General basic probabilities (see Figure EG4.1, below). The work-
Addition Rule
sheet contains the Table 4.1 purchase behavior data shown
PHStat2 Use Simple & Joint Probabilities to compute on page. Overwrite these values when you enter data for
basic probabilities. Select PHStat Probability & Prob. other problems.
Distributions Simple & Joint Probabilities. The proce- Open to the COMPUTE_FORMULAS worksheet to
dure inserts a worksheet similar to Figure EG4.1 into the examine the formulas used in the worksheet, many of which
current workbook. (Unlike with other procedures, no dialog are shown in the inset to Figure EG4.1.
box is first displayed.) To use the worksheet, fill in the
Sample Space area with your data.
FIGURE EG4.1 COMPUTE worksheet of the Probabilities workbook
EG4.2 CONDITIONAL PROBABILITY

There is no Excel material for this section.
EG4.3 BAYES THEOREM

In-Depth Excel Use the COMPUTE worksheet of the
Bayes workbook as a template for computing basic
probabilities (see Figure EG4.2, at right). The worksheet
contains the television-marketing example of Table 4.4 on
page 164. Overwrite these values when you enter data for
other problems.
Open to the COMPUTE_FORMULAS worksheet to
examine the simple arithmetic formulas that compute the FIGURE EG4.2 COMPUTE worksheet of the
probabilities which are also shown in the inset to Figure EG4.2. Bayes workbook
177
EG4.4 COUNTING RULES can be arranged. For example, the formula =FACT(6)
computes 6!
Counting Rule 1
In-Depth Excel Use the POWER(k, n) worksheet function Counting Rule 4
in a cell formula to compute the number of outcomes given In-Depth Excel Use the PERMUT(n, x) worksheet func-
k events and n trials. For example, the formula =POWER(6, tion in a cell formula to compute the number of ways of
2) computes the answer for Example 4.11 on page 168. arranging x objects selected from n objects in order. For
example, the formula =PERMUT(6, 4) computes the answer
Counting Rule 2 for Example 4.14 on page 169.
In-Depth Excel Use a formula that takes the product of Counting Rule 5
successive POWER(k, n) functions to solve problems re-
lated to counting rule 2. For example, the formula In-Depth Excel Use the COMBIN(n, x) worksheet func-
=POWER(26, 3) * POWER(10, 3) computes the answer for tion in a cell formula to compute the number of ways of
the state motor vehicle department example on page 168. arranging x objects selected from n objects, irrespective of
order. For example, the formula =COMBIN(6, 4) computes
the answer for Example 4.15 on page 170.
Counting Rule 3
In-Depth Excel Use the FACT(n) worksheet function
in a cell formula to compute how many ways n items
C H A P T E R 4 M I N I TA B G U I D E
MG4.1 BASIC PROBABILITY CONCEPTS 3. Click OK.

There is no Minitab material for this section. If you have previously used the Calculator during your
Minitab session, you may have to clear the contents of the
Expression box by selecting the contents and pressing Del
MG4.2 CONDITIONAL PROBABILITY before you begin step 2.
There is no Minitab material for this section.
MG4.3 BAYES THEOREM

There is no Minitab material for this section.
MG4.4 COUNTING RULES

Use Calculator to apply the counting rules. Select Calc
Calculator. In the Calculator dialog box (shown at right):
1. Enter the column name of an empty column in the Store
result in variable box and then press Tab.
2. Build the appropriate expression (as discussed later in this
section) in the Expression box. To apply counting rules 3
through 5, select Arithmetic from the Functions drop-
down list to facilitate the function selection.
Chapter 4 Minitab Guide 179
Counting Rule 1 Counting Rule 4

Enter an expression that uses the exponential operator **. Enter an expression that uses the PERMUTATIONS(n, x)
For example, the expression 6 ** 2 computes the answer for function to compute the number of ways of arranging x
Example 4.11 on page 168. objects selected from n objects in order. For example, the
expression PERMUTATIONS(6, 4) computes the answer
Counting Rule 2 for Example 4.14 on page 169.
Enter an expression that uses the exponential operator **.
For example, the expression 26 ** 3 * 10 ** 3 computes the Counting Rule 5
answer for the state motor vehicle department example on Enter an expression that uses the COMBINATIONS(n, x)
page 168. function to compute the number of ways of arranging x
objects selected from n objects, irrespective of order. For
Counting Rule 3 example, the expression COMBINATIONS(6, 4) computes
Enter an expression that uses the FACTORIAL(n) function the answer for Example 4.15 on page 170.
to compute how many ways n items can be arranged. For
example, the expression FACTORIAL(6) computes 6!

Chapter 4 Basic Probability - 2 PDF

Uploaded by

Copyright:

Available Formats

Chapter 4 Basic Probability - 2 PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Chapter 4 Basic Probability - 2 PDF

Uploaded by

Copyright:

Available Formats

4 Basic Probability

USING STATISTICS @ M&R Decision Trees Counting Rule 3

4.2 Conditional Probability 4.4 Counting Rules CHAPTER 4 MINITAB GUIDE

Basic probability concepts

he principles of probability help bridge the worlds of descriptive statistics and

4.1 Basic Probability Concepts

X = number of ways in which the event occurs

Events and Sample Spaces

Contingency Tables and Venn Diagrams

FIGURE 4.1 B FIGURE 4.2 A B A B = 650

SOLUTION Using the following definitions:

Planned to purchase and actually purchased

P(Planned to purchase) = P(Planned to purchase and purchased)

General Addition Rule

P(Planned to purchase or actually purchased) = P(Planned to purchase and did

GENERAL ADDITION RULE

P(Planned to purchase or actually purchased) = P(Planned to purchase)

Problems for Section 4.1

AGE OF RESPONDENTS GENDER

4.2 Conditional Probability

Computing Conditional Probabilities

Planned to purchase and actually purchased

You can also use Equation (4.4b) to compute this result:

Example 4.6 further illustrates conditional probability.

Number purchasing television with

If you use Equation (4.4b) on page 155:

P1A and B2 38>300

Example 4.7 illustrates how to construct a decision tree.

Example 4.8 demonstrates the use of Equation (4.5).

TA B L E 4 . 3 SATISFIED WITH PURCHASE?

and solving for the joint probability P(A and B).

GENERAL MULTIPLICATION RULE

Example 4.9 demonstrates the use of the general multiplication rule.

P1A and B2 = a b a b = 0.6380

MULTIPLICATION RULE FOR INDEPENDENT EVENTS

Marginal Probability Using the General Multiplication Rule

MARGINAL PROBABILITY USING THE GENERAL MULTIPLICATION RULE

P1A2 = P1A|B12P1B12 + P1A|B22P1B22 + + P1A|Bk2P1Bk2 (4.8)

To illustrate Equation (4.8), refer to Table 4.1 on page. Let

Problems for Section 4.2

NEED THREE OR MORE CLICKS

a. Given that a worker is from the United States, what is the

4.3 Bayes Theorem

To use Equation (4.9) for the television-marketing example, let

Then, using Equation (4.9),

TA B L E 4 . 4 Prior Conditional Joint

P(S and F) = P(F|S) P(S)

P(S and F) = P(F|S) P(S)

P(S and F) = P(F|S) P(S)

Example 4.10 applies Bayes theorem to a medical diagnosis problem.

TA B L E 4 . 5 Prior Conditional Joint

P(D and T ) = P(T |D) P(D)

P(D and T ) = P(T |D) P(D)

P(D and T ) = P(T |D) P(D)

THINK ABOUT THIS Divine Providence and Spam

Problems for Section 4.3

4.4 Counting Rules

COUNTING RULE 4: PERMUTATIONS

COUNTING RULE 5: COMBINATIONS