CHAPTER TWO: Probability

Definitions
An event is a specified result that may or may not occur when an experiment is performed.

Examples: Experiment Event .

a coin is to be flipped “heads” comes up
two dice are to be rolled “doubles” comes up
a roulette wheel is spun “00” comes up
a person is randomly selected that person has Type A blood

The probability of an event is a numerical measure of the likelihood of its occurrence or truth. It is a number between
0 and 1, inclusive.
Probability = 1 means the event must occur
Probability = 0 means the event cannot occur
A natural way to quantify likelihood is to consider the proportion of times that the event would occur if the experiment
or situation is repeated over and over many times. Equate the probability with the long run proportion of occurrences.
This may be the only way to determine likelihood; we have to take the time to repeat the experiment many, many
times or look at past records.

But it is often the case that we can side-step all that work and arrive at probabilities by making certain assumptions
about the problem at hand. For example, in order to determine the probability that “heads” comes up on the flip of
a brand new coin, we need not flip the coin 10,000 times …. we can simply look at the coin.

Given no physical difference between the two sides, you have no reason to believe that heads is more or less likely
than tails – hence probabilities of ½ and ½ respectively.

Such assumptions constitute what is known as a probability model.

For the most part, we will look at experiments and situations in which the possible outcomes are equally likely – in
which case probabilities are easy to find:

Suppose that an experiment or situation has N possible outcomes. If the outcomes are assumed to be equally
likely, then the probability that a particular event occurs

the total number of ways that the event can occur

= -------------------------------------------------------------------
N

NOTE: The “equal likelihood” assumption often applies. Consider that many experiments are designed in such a
way that the outcomes will be equally likely (i.e. simple random sampling).

PROBLEM 2.1
A class has eight students; the grade (1,2,3,4) is given for each.
A student is to be selected at random.

What is the probability that …..

(a) a male is selected?
(b) a 2nd year student is selected?
(c) a 4th year student is not selected?
(d) a 1st year male is selected?

13
PROBLEM 2.2
Suppose that a playing card is to be randomly selected from a full
deck of fifty-two. (All 52 cards are pictured to the right.)
Find the probability of selecting:
(a) the ace of spades
(b) a face card (J, Q, K)
(c) a red card
(d) a card that is not a king

PROBLEM 2.3
A fair coin will be flipped four times; the equally likely possible sequences are listed below.

What is the probability that

(a) exactly three heads are flipped?
(b) at least two heads are flipped?
(c) you get at least three tails in a row?
Now suppose instead that the same coin will be flipped ten times in a row.
(d) What is the probability of flipping ten straight heads?

Probability and Odds

If you know the probability of an event, you can easily compute the odds that it occurs (and vice versa):

𝟏−𝐩
Probability = p → "( ) 𝐭𝐨 𝟏 𝐨𝐝𝐝𝐬"
𝐩

𝟏
"K to 1 odds" → 𝐏𝐫𝐨𝐛𝐚𝐛𝐢𝐥𝐢𝐭𝐲 =
𝐊+𝟏

At this point we need to give some mathematical structure to the problem of finding probabilities.

Definitions
The sample space for an experiment or situation is the collection of all possible outcomes.
An event is any collection of outcomes (i.e. a subset of the sample space).

For the sake of convenience (and to keep our soon-to-be developed probability rules simple), events are denoted by
capital letters: A , B , C , E , M , A1 , A2 , ….. (the letter S is reserved to denote the sample space). Now an event
can be described both in words and in set notation.

Example
A fair coin is flipped four times.
In words: Let W be the event that exactly two heads are flipped
Let L be the event that the first flip is tails
In set notation: W = {HHTT, HTHT, HTTH, THHT, THTH, TTHH}
L = {THHH, THHT, THTH, THTT, TTHH, TTHT, TTTH, TTTT}

Example
A student is picked from the class in Problem 7.
In words: Let F be the event that a female is selected
Let G2 be the event that a 2nd year student is selected
In set notation: F = {Mary, Kathy, Diane, Judy, Debbie}
G2 = {Steve, Kathy}

14
Definition
A Venn Diagram is a graphical display of the sample space and various events of interest.
Venn Diagrams will help us better understand how various events relate to each other.

Relationships between events

o complement
o intersection
o union

Let E be an event. Then the collection of outcomes not belonging to E is also an event. We call this the complement
of E and denote it by E.

Example
For the class in Problem 2.1, where
M is the event that a male is selected
G1 is the event that a first year student is selected

M = {Debbie, Kathy, Mary, Judy, Diane}

G1 = {Debbie, Steve, Kathy}

Now let A and B be a pair of events. The collection of outcomes belonging to both A and B is also an event. We call
this the intersection of A and B and denote it by (A & B).

Example
For the class in Problem 2.1, where
M is the event that a male is selected
G1 is the event that a first year student is selected

(M & G1) = {John, Fred}

Again let A and B be a pair of events. The collection of outcomes belonging to at least one of A and B is also an
event. We call this the union of A and B and denote it by (A or B). This is the event that either A or B occurs.

Example
For the class in Problem 2.1, where
M is the event that a male is selected
G1 is the event that a first year student is selected

(M or G1) = {John, Fred, Steve, Mary, Judy, Diane}

The concepts extend to groups of events; for example, suppose that A, B, C, and D are events:

Intersection of the four events Union of the four events

(A & B & C & D) (A or B or C or D)

(the outcomes that belong to all) (the outcomes that belong to at least one)

15
We need some more notation.

For some event E, let n(E) be the total number of possible outcomes in E
P(E) be the probability of E.

If all of the possible outcomes are assumed to be equally likely, then

𝐧(𝐄)
𝐏(𝐄) =
𝐧 (𝐒 )
Loosely speaking, the union of two events is additive.

if A and B do have common outcomes: P(A or B) = P(A) + P(B) – P(A & B)

if A and B do not* have common outcomes: P(A or B) = P(A) + P(B)
Definition
If A and B do not share any common outcomes, then we say they are mutually exclusive.

Example
For the problem of rolling two balanced dice, where ● ●
●
●
●
● ●
● ●
●
●
●
●
●
●
●
●
● ●

• V is the event that the sum is equal to 8

● ● ● ● ● ● ● ●

● ● ● ● ● ● ● ● ● ● ● ● ● ●

• W is the event that the product is equal to 12 ●

● ● ●
●
● ● ● ● ● ●
●
● ●
●
●
●
● ●

● ● ● ● ● ● ● ● ● ● ● ● ● ●
● ● ● ● ● ● ● ● ● ● ●

n(V) = 5 P(V) = 5/36 = .139

● ● ● ● ● ● ● ● ● ● ● ● ● ●

● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●
n(W) = 4 P(W) = 4/36 = .111 ●
● ● ● ● ●
●
● ● ● ● ● ● ● ●
●
● ● ●
●
●
●
● ● ●

● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●
● ● ● ● ● ● ● ● ● ● ●

n(V & W) = 2 P(V & W) = 2/36 = .056 ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●

● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●
n(V or W) = 7 ● ●
●
●
● ●
●
●
●
●
●
●
●
●
●
● ● ●
●
●
●
● ●
●
●
●
●
●
●
●
●
●
●
●
●
●
●

By the formula above: P(V or W) = [ P(V) + P(W) – P(V & W) ]

= 5/36 + 4/36 – 2/36 = 7/36 = .194

2x2 probability tables

Every pair of events (call them A and B) generate a set of four intersections that cover the entire sample space:

• (𝐴 & 𝐵) "both A and B occur" A&B

• (𝐴 & 𝐵̅) "A occurs but B does not occur" A&B

A&B A&B

• (𝐴̅ & 𝐵) "B occurs but A does not occur"

• (𝐴̅ & 𝐵̅) "neither A nor B occurs"

16
Probabilities can then be summarized in a 2x2 probability table.

Note that the sample space is composed of the four inner cells.

PROBLEM 2.4
Among the high schools in a large county
• 82% have a baseball field on campus
• 71% have a football field on campus
• 68% have a baseball and a football field on campus
What percentage of schools in this district:
(a) do not have a football field on campus?
(b) have a baseball field on campus but not a football field?
(c) have neither a baseball field nor football field on campus?
(d) have either a baseball field or a football field on campus
(or both)?

HINT: Fill out the 2x2 probability distribution at the right.

Probability Rules

We will examine and apply:

1. Complementation Rule
2. Conditional Probability Rule
3. General Multiplication Rule
4. Special Multiplication Rule (which only works for independent events)

PROBABILITY RULE #1: Complementation Rule

For any event E

𝐏(𝐄) = 𝟏 − 𝐏(𝐄̅)
(Or in other words: The probability that something happens is 1 minus the probability that it doesn’t happen!)

17
Conditional Probability

Occasionally, some (but not all) information about the outcome of an experiment is learned. This can cause some
events to be more or less likely. How?

PROBLEM 2.5
A student is to be selected from the class mentioned in Problem 2.1.
(a) What is the probability that the selected student is a 2nd year student?
(b) Suppose you obtain the information that the selected student is a male.
(In other words, you know that the event M has occurred.)
What is the (new) probability that the selected student is a 2nd year student?

Definition
Suppose A and B are events with respect to some experiment. The probability of B given the information that A has
occurred is called a conditional probability and is denoted by P(B│A).

 the unconditional probability of B is P(B) (given NO information)

 the conditional probability of B is P(B│A) (given the information that A has occurred)

If we learn that A has occurred, then we can “update” the sample space by eliminating all of the once-possible
outcomes that are not in A. Then we find the probability of B in the usual manner …. count the number of ways
that B can still occur and divide by the “new” total number of outcomes.

The “new” sample space is merely all of A! The only part of B that remains is whatever part overlaps with A.
Therefore:

𝐧(𝐀&𝐁)
𝐏(𝐁|𝐀) =
𝐧(𝐀)
PROBLEM 2.6
A single playing card is to be randomly selected from a deck of 52.
Let Q be the event that a queen is selected
Let D be the event that diamonds is selected
Let R be the event that a red card is selected
Let F be the event that a face card (J, Q, K) is selected
Let E be the event that the card is even numbered (2, 4, 6, 8, 10)
Although you do not know the exact card that was selected, some information may be given to you. (Imagine that
you pick the card blindfolded and hand it to your friend, who may or may not tell you something about it.) Find:
(a) P(Q) (e) P(Q│D)
(b) P(D) (f) P(Q│F)
(c) P(R) (g) P(R│D)
(d) P(F) (h) P(Q│(R & F))

18
We have been finding conditional probabilities directly (i.e. by updating the sample space, etc.). We can also
express a conditional probability as the ratio of two unconditional probabilities.

PROBABILITY RULE #2: Conditional Probability Rule

For any two events A and B,

𝐏(𝐀&𝐁)
𝐏(𝐁|𝐀) =
𝐏(𝐀)
PROBLEM 2.7
Of all members of the 112th House of Representatives:
• 55.6% are Republicans
• 21.4% are freshmen (i.e. “first-termers”)
• 19.3% are freshmen Republicans
A member of the House is selected at random.
(a) Given the information that a Republican was selected, what is the probability that he/she a freshman?
(b) Given the information that a freshman was selected, what is the probability that he/she is a Republican?
(c) Given the information that a non-freshman was selected, what is the probability that he/she is not a
Republican?

Depending on what probabilities are already known, we can use the Conditional Probability Rule to find certain
unconditional probabilities. (The rule consists of three terms … as long as we know two of them we can solve for
the third.) By multiplying both sides of the CPR by P(A), we get …..

PROBABILITY RULE #3: General Multiplication Rule

For any two events A and B:

𝐏(𝐀&𝐁) = 𝐏(𝐀)𝐏(𝐁|𝐀)
Rule #3 is especially useful for experiments that consists of two random selections, one at a time, where A1 is an
event that is concerned solely with the first selection and A2 is an event that is concerned with the second selection.

The probability that A1 and A2 both occur in succession is therefore: P(A1 & A2) = P(A1) P(A2│A1)
Remember that multiple selections can be carried out in two ways: with or without replacement.

With replacement: allows for multiple selections of the same person or thing
Without replacement: once a person or thing is selected it cannot be chosen again

PROBLEM 2.8
Consider the class from Problem 2.1. Two students will be randomly selected,
one at a time, without replacement. Find the probability that:
(a) both picks are females
(b) the 1st pick is a female and the second pick is a male
(c) the selected students are the same gender
(d) exactly one female is selected
(e) at least one female is selected

19
PROBLEM 2.9
Suppose that two playing cards are randomly selected, one at a time, without replacement.
(a) What is the probability that both cards are “clubs”?
(b) What is the probability that both cards are the same suit?
(c) What is the probability that both cards are the same denomination?
(d) What is the probability that you get “21”?

Let’s return to an experiment that consists of a single selection from a population.

Suppose that one card is selected from a full deck, where:

Q is the event that a queen is picked
D is the event that diamonds is picked
F is the event that a face card (J, Q, K) is picked

We already determined that:

P(Q) = 4/52 = .077
P(Q│D) = 1/13 = .077
P(Q│F) = 4/12 = .333

Knowledge that the selected card was “diamonds” did not change the probability of “queen” …. knowledge that
the selected card is a face card, on the other hand, does change the probability of “queen”. This illustrates yet
another important concept that pertains to pairs (and groups) of events.

Definition
Two events A and B are said to be independent if the known occurrence of one does not change the probability that
the other occurs, in which case

P(A) = P(A│B) and P(B) = P(B│A)

Otherwise, A and B are said to be dependent.

In order to check for independence, you only need to verify that one of the above equations is true. They are either
both true, or they are both false.

Independence is often a “common sense” assumption that we make, just like assuming that outcomes are equally
likely. Consider the following events:

• Let P be the event that the Phillies (baseball) reach the playoffs in the year 2024
• Let J be the event that the Dow Jones Industrial Average increases by at least 10% during 2024
• Let M be the event that the Mets (baseball) reach the playoffs in 2024
• Let D be the event that a Democrat is elected President in 2024

Can you identify pairs of independent events among them?

Sometimes it is not immediately clear whether we can assume independence; we have to use the definition to
verify.

PROBLEM 2.10
A joint frequency distribution is given for the workers in a particular office, grouped according to whether they are
male or female and whether or not they have siblings. A worker is to be randomly selected.

Let M be the event that a male is selected

Let O be the event that a person is an “only child”.

Are M and O independent?

20
Recall Rule #3: P(A & B) = P(A) P(B|A). This is true for all pairs of events. How does this equation simplify in the
special case that A and B are independent?

PROBABILITY RULE #4: Special Multiplication Rule

For independent events A and B

𝐏(𝐀&𝐁) = 𝐏(𝐀)𝐏(𝐁)
The above rule can be used to determine if events are independent or dependent.

PROBLEM 2.11
Of the homes in a small town, 60% have basements and 25% have swimming pools. Furthermore, 18% have both
basements and swimming pools. For a randomly selected home:

Let B be the event that the home has a basement

Let W be the event that the home has a swimming pool

Are B and W independent events? Why or why not?

Above, we are given probabilities and asked to determine if events are independent.
We often assume independence in order to find probabilities.

PROBLEM 2.12
A pair of balanced dice are to be rolled twice.
Let N1 be the event that the first roll is “snake-eyes”
Let N2 be the event that the second roll is “snake-eyes”
Assuming that N1 and N2 are independent (why?), what is the probability that both rolls result in “snake-eyes”?

The concept of independence generalizes to groups of events.

Definition
Events A, B, C, D, … are said to be independent (as a group) if no two of them are dependent, in which case:

P(A & B & C & D & … ) = P(A) P(B) P(C) P(D)    

The above formula is very, very, very important. It allows us to find the probability that a whole chain of events
occur, one after another, whenever the events are all independent of each other. Just multiply right down the line!

PROBLEM 2.13
What is the probability of rolling five “snake-eyes” in a row (with balanced dice)? What are the odds?

Independent Trials

Many experiments consist of repeated, identical “mini-experiments” or trials. We say that the trials are
independent if all of the events that pertain to any one trial are independent of all of the events that pertain to all of
the other trials. In other words, they are independent when whatever happens on one trial does not change
probabilities for all of the other trials.

(Examples of experiments that consist of independent trials: dice rolls, coin flips, spins of a wheel)

21
PROBLEM 2.14
The casino game of roulette results in a single randomly selected outcome among thirty-eight possibilities on a
wheel. The possibilities are both numbered and colored.

Among the 38 possibilities:

18 of them are red (prob. of red = 18/38 = .474)
18 of them are black (prob. of black = 18/38 = .474)
2 of them are green (prob. of green = 2/38 = .053)

Suppose you play five games in a row.

(a) What is the probability that you get red outcomes all five times?
(b) What is the probability that the first spin results in green, the next
three spins result in black, and the last spin is red?
(c) What is the probability that none of the five spins result in a green outcome?
(d) What is the probability that at least one of the five spins results in a green outcome?

Q: But what about the more important problem involving repeated selections from a population (i.e. simple
random sampling)? Do such selections constitute independent trials?

A: If the selecting is done with replacement, the selections are independent

If the selecting is done without replacement, the selections are NOT independent

Now consider the following problem.

PROBLEM 2.15
The Springfield County Chamber of Commerce has 375 members, of which 108 own their own business. (That’s
exactly 28.8%.) Suppose that two members are selected at random. Find the probability that both of the selected
members are business owners, if the selecting is done
(a) with replacement
(b) without replacement
What do you notice about your answers? How can you explain it?

If the number of without-replacement selections (a sample of n of them) is “small” relative to the size of the
population (of size N), then we can say that the selections are almost independent … we treat the selections as if they
are independent, in order to get very good approximation to the actual probabilities.

RULE OF THUMB: If the sample size is less than 5% of the population, go ahead and treat the selections as if they
are independent. (i.e. if n/N < .05). It will often be very obvious that this rule of thumb is met.

PROBLEM 2.16
A recent Gallup poll estimates that 58% of all American adults own stocks. Assume this is true. For a random
sample of four American adults, find the probability that
(a) all four own stocks
(b) none of the four own stocks
(c) at least one of the four owns stocks
(d) exactly one of the four owns stocks
(e) exactly two of the four owns stocks (This one is difficult now, but will be made easier in the next chapter.)

22
Lotteries

PROBLEM 2.17
Suppose that a lottery game consists of choosing three digits, each one from 0 to 9. (Your number must be between
000 and 999, inclusive.) These numbers will be matched against three numbered ping-pong balls, each one
randomly selected from its own container. Each container holds balls numbered 0-9.
(a) On any given day, what is the probability of winning this lottery?
(b) Suppose you play this game every day for a year. What is the probability of winning at least once?
(a) What is the probability of winning at least twice in a year?

PROBLEM 2.18
Now suppose that a lottery game involves choosing six digits, each one from 0 to 9. (Your number must be
between 000000 and 999999, inclusive.) These numbers will be matched against six numbered ping-pong balls, each
one randomly selected from its own container. Each container holds balls numbered 0-9.
(a) On any given day, what is the probability of winning this lottery?
(b) Suppose you play this game every day for a year. What is the probability of winning at least once?
(c) Suppose you play this game every day of the year for fifty years.
What is the probability of winning at least once?

PROBLEM 21.9 (The New Jersey Pick-6 Lotto)

Now consider a lottery game that involves choosing six different numbers among those from 1 to 49. The numbers
cannot repeat. Six ping-pong balls are chosen without replacement from a container of balls numbered 1 to 49. You
win if all of your numbers match up in any order.
(a) On any given day, what is the probability that you win this lottery?
(b) If you play this game every day for fifty years, what is the probability you win at least once?
(c) If you were to play every day for one thousand years, what is the probability you win at least once?
(d) In order for the probability of winning at least once to be greater than ½, how many days must this game
be played?

23