STAT201 Elementary Statistical Inference

Chapter 3: Probability

Prasenjit Ghosh

1 / 90
Terminology

Random Experiment
An experiment is a planned operation carried out under controlled
conditions. If the result or outcome of an experiment is not
predetermined, that is, cannot be predicted with certainty before the
experiment is performed, the experiment is called a chance or a
random experiment.

Examples
Flipping a coin
Rolling a fair die
Lifetime of an bulb or an electronic device
Number of calls to a communication device
Weather condition tomorrow (whether it will rain or not)

2 / 90
Events
The result or outcome of a random experiment is called an event.
Upper case letters like A, B etc are used to denote events.
If an event cannot be disintegrated further into more than one
events, it is called an elementary event.
If an event can be written as a composition of more than one
events, it is called a composite event.
The sample space S of a random experiment is the set of all possible
elementary events.

Examples
1 Flipping a fair coin once. There are only two possibilities: either a
head (H) or a tail (T). The sample space is S = {H, T}.
2 Flipping a fair coin twice. Here, S = {HH, HT, TH, TT}. ”Resulting
a head in the first flip”= {HH, HT} is a composite event here.
3 / 90
Events and Sample Space - More Examples

Example
The event of observing a vowel is {a, e, i, o, u} when a letter is
drawn at random from the English alphabet is composite.

The event of choosing an odd integer between 1 and 9 is

{1, 3, 5, 7, 9} is composite. Here, S = {1, 2, ... , 9}.

Rolling a fair die once. Here, S = {1, 2, 3, 4, 5, 6}. The event of

getting an even number {2, 4, 6} is composite.

The event of at least two successful surgeries {SSF, FSS, SFS,

SSS} is a composite event when a surgeon operates three
patients in a day. Here, S = {SSS, SSF, FSS, SFS, SFF, FSF,
FFS, FFF}, with S and F denoting respectively the success or
failure of a surgery.

4 / 90
Set Operations
The union A ∪ B of events A and B is the event consisting of all
outcomes that are in A or in B or in both (also written A OR B)

The intersection A ∩ B of A and B is the event consisting of all

outcomes that are in both A and B (also written A AND B)

The complement Ac of A is the event consisting of all outcomes

that are not in A (also written APime )

The events A and B are said to be mutually exclusive or disjoint

if they have no outcomes in common, that is, if A ∩ B = ∅, where ∅
denotes the empty (or, the null) set

The difference A − B is defined as A ∩ Bc

A is a subset of B, A ⊂ B (if e ∈ A implies e ∈ B)

Two sets are equal, A = B, if A ⊂ B and B ⊂ A.

5 / 90
Source: https://www.onlinemathlearning.com/union-set.html
6 / 90
Set Operations: Example

Example
Let us draw a digit at random from the set of all possible digits
{0, 1, 2, 3, 4, 5, 6, 7, 8, 9}. Here, the sample space

S = {0, 1, 2, 3, 4, 5, 6, 7, 8, 9}.

Define the following events:

A = {0, 1, 2, 3, 4, 5}
B = {3, 4, 5, 6, 7, 8}
C = {1, 3, 5, 7, 9}

Find

Ac , A ∩ B, A ∩ C, A ∪ B, A ∪ C, A − B, B − A, A − C, C − A, B − C, C − B.

7 / 90
Set Operations: Example
Example
A vehicle taking a particular freeway exit can turn right (R), turn left (L),
or go straight (S)
Random experiment: the direction for three successive vehicles
What does one of the outcomes in the sample look like?

All outcomes in the event

A = {all three vehicles go in the same direction}

List all outcomes in the event

B = {exactly two vehicles turn right}

List all outcomes in the event

C = {exactly two vehicles go in the same direction}

List outcomes in Ac and A ∩ B

8 / 90
Definition of Probability
Probability is a measure or ”size” associated with the outcomes of a
random experiment that quantifies how likely a given outcome is to
occur when the experiment is performed.
We write the probability of an event A as P(A) or Pr(A).
The definition of probability must satisfy these three axioms or
fundamental principles:

Axiom 1: 0 ≤ P(A) ≤ 1, for all events A

Axiom 2: P(S) = 1
Axiom 3: If A1 , A2 , ... are disjoint

P(A1 ∪ A2 ∪ · · · ) = P(A1 ) + P(A2 ) + · · ·

The axioms imply: P(∅) = 0 where ∅ denotes the null (or, the
improbable) event.
9 / 90
Probability Laws
Rule 1: For any event A, P(Ac ) = 1 − P(A)
Rule 2: For any two events A and B
P(A OR B) = P(A) + P(B) − P(A AND B).
(Heuristically, A AND B is taken into account twice while computing
P(A OR B). So, we need to subtract P(A AND B) from P(A) + P(B) as an
adjustment to obtain P(A OR B).)
Rule 3: If A and B are mutually exclusive
P(A AND B) = 0.
Rule 4: If A and B are mutually exclusive
P(A OR B) = P(A) + P(B).
Rule 5: P(A) = P(A AND B) + P(A AND Bc ).
Rule 6: If A ⊂ B, P(A) ≤ P(B).
10 / 90
How to Calculate Probabilities

The Classical Approach

It is possible for all possible elementary outcomes of a random
experiment to be equally likely
That is, each elementary outcome has an equal chance of
occurrence.

Examples
If you toss a fair coin once, a head (H) and a tail (T) are equally
likely to occur.
If you toss a fair, six-sided die once, each face (1, 2, 3, 4, 5, or 6)
is as likely to occur as any other face.
If you randomly guess the answer to a True/False question on an
exam, (typically) you are equally likely to select a correct answer
or an incorrect answer.
11 / 90
Classical Definition of Probability

To calculate the probability of an event A using the classical definition,

assume that all the elementary events in the sample space S are
equally likely.
1 Count the total number of elements in S
2 Count the number of elements in S that favors the occurrence of A
3 Divide the count in Step (2) by the count in Step (1)

Thus
Number of ways A occurs
P(A) =
Number of elements in the sample space S

12 / 90
Examples

A jar of 150 jelly beans contains 22 red jelly beans, 38 yellow, 20

green, 28 purple, 26 blue, and the rest are orange. A single jelly bean
is drawn at random from the jar. Here, the sample space S is the
collection of all 150 jelly beans in the jar.
Let B be the event of drawing a blue jelly bean. Then, by the classical
definition, P(B) = 26/150 = 13/75.

There are 23 countries in North America, 12 countries in South

America, 47 countries in Europe, 44 countries in Asia, 54 countries in
Africa, and 14 in Oceania (Pacific Ocean region).
Let A be the event that a randomly selected country is in Asia. Here,
the sample space S is the collection of all the 204 countries in the
world. Then, assuming each country is equally likely to be drawn, by
the classical definition, P(A) = 44/204 = 11/51.
13 / 90
Examples

Suppose a letter is drawn at random from the English alphabet. Then

the sample space S would be the English alphabet containing 26 many
equally likely elementary outcomes (that is, all 26 letters in the English
alphabet).
Let A = {a, e, i, o, u} be the event that the letter drawn is a vowel.
Hence, by the classical definition, P(A) = 5/26.

Suppose a number is drawn at random from the set of first 100 natural
numbers (positive integers). Here, the sample space S is given by

S = {1, 2, 3, ... , 99, 100}.

Let A = {5, 10, 15, ... , 95, 100} be the event that the number drawn is a
multiple of 5. Hence, by the classical definition, P(A) = 20/100 = 1/5.

14 / 90
Coin Tossing Examples
Suppose you toss a fair coin twice. Here, the sample space S has four
equally likely elementary events:

S = {HH, TH, HT, TT}

Define the event A as ”getting exactly one head”. Then A = {HT, TH}.
Hence, by the classical definition P(A) = 24 = 0.5.

Suppose you toss a fair coin thrice. Here, the sample space S has
eight equally likely elementary events:

S = {HHH, HHT, HTH, THH, HTT, THT, TTH, TTT}

Define the event A as ”getting a head at the second flip”. Then

A = {HHH, HHT, THH, THT}. Hence, by the classical definition,
P(A) = 48 = 0.5.
15 / 90
Die Throwing Example
Suppose you roll a fair six-sided die, with the numbers 1, 2, 3, 4, 5, 6 on
its faces. Here, the sample space S = {1, 2, 3, 4, 5, 6} has six equally
likely elementary events.

Define the event A as ”getting an even number”. Then A = {2, 4, 6}.

Hence, by the classical definition, P(A) = 63 = 0.5.

Define the event B as ”getting at least five”. Then B = {5, 6}. Hence,
by the classical definition, P(B) = 26 = 31 .

P(A AND B) = 16 .
4
P(A OR B) = 6 = 23 .
2 1
P(A – B) = 6 = 3.
1
P(B – A) = 6.
16 / 90
Throwing of a Fair Die Twice

Suppose we roll a fair die twice. The sample space would comprise of
6 × 6 = 36 many paired outcomes as follows:

(1, 1), (1, 2), (1, 3), (1, 4), (1, 5), (1, 6),
 

 
 


(2, 1), (2, 2), (2, 3), (2, 4), (2, 5), (2, 6),



 
(3, 1), (3, 2), (3, 3), (3, 4), (3, 5), (3, 6),

S=


(4, 1), (4, 2), (4, 3), (4, 4), (4, 5), (4, 6),


 



(5, 1), (5, 2), (5, 3), (5, 4), (5, 5), (5, 6),



 
(6, 1), (6, 2), (6, 3), (6, 4), (6, 5), (6, 6)
1
Note that the probability of each outcome in S is 36 .

Let A be the event that the sum of the two faces equals to 7. Then,
6
A = {(1, 6), (2, 5), (3, 4), (4, 3), (5, 2), (6, 1)}. Hence, P(A) = 36 = 16 .
17 / 90
Continuation of the Preceding Example
Let B be the event that the sum of two faces is an even number (i.e. a
multiple of 2). Then
 
(1, 1), (1, 3), (1, 5), (2, 2), (2, 4), (2, 6),
 
B = (3, 1), (3, 3), (3, 5), (4, 2), (4, 4), (4, 6),
 
(5, 1), (5, 3), (5, 5), (6, 2), (6, 4), (6, 6)
 

18
Hence, by the classical definition P(B) = 36 = 12 .

Let C be the event that the sum of two faces is a multiple of 3. Then
( )
(1, 2), (1, 5), (2, 1), (2, 4), (3, 3), (3, 6),
C=
(4, 2), (4, 5), (5, 1), (5, 4), (6, 3), (6, 6)

12
Hence, by the classical definition P(C) = 36 = 31 .
18 / 90
Card Drawing Example
Suppose a card is drawn at random from a well-shuffled deck of 52
cards:

Let A and B be the events that the card drawn is a face card (that
means, it is either ”J”, ”Q” or ”K”), and a spade, respectively. Then,
12 3 13 1
P(A) = = , and P(B) = = .
52 13 52 4
19 / 90
Shortcomings of the Classical Definition
1 The definition is circular in nature.
2 Equally likely elementary events are rare in practice.
3 S may have an infinitely or uncountably many elements.

Examples
The probabilities of a head (H) and a tail (T) when a biased coin is
thrown once are not the same.
Occurrence of a six may not be as likely as the occurrence of the
other faces when you throw a loaded die.
The success probability of a particular knee surgery when two
surgeries are done in a day is 0.85. Here, S = {SS, SF, FS, FF}.
However, the elements of S need not be equally likely. Hence, the
probability of two successful surgeries is not 41 in this case.
The probability that a randomly chosen number from (0, 1) will lie
in (0.2, 0.5) can’t be determined using the classical definition. 20 / 90
Conditional Probability – Motivation
A card is drawn randomly from a well–shuffled deck of 52 cards:

What is the (unconditional) probability the card is a king?

Suppose now you are told that it is a face card. Given this
information
4
what is the probability the card is a king? (Ans: 12 = 31 )
what is probability that it is a 2? (Ans: 0)
21 / 90
Conditional Probability – Definition
Definition
The conditional probability of A given B, denoted P(A|B), is the
probability that event A will occur given that the event B has already
occurred, and is defined as
P(A AND B)
P(A | B) =
P(B)

provided P(B) > 0 (otherwise this definition becomes invalid).

Example
In the preceding example, let A and B denote the events ”the card
drawn is a King” and ”the card drawn is a face card”, respectively.
4 12
Here A ⊂ B. So, P(A AND B) = P(A) = 52 . Also, P(B) = 52 .
P(A AND B)
Hence, P(A | B) = P(B) = (4/52)/(12/52) = 1/3.
22 / 90
Two Important Notes

Reduced Sample Space

A conditional reduces the sample space.
Given B, identify those elements of B which satisfy the occurrence
of A (that is, identify the elements of A AND B).
In the preceding example, when it is given that you have already
drawn a face card, confine your attention only to face cards now.
Thus, given the aforesaid information, the conditional probability
4
that the card drawn is a king is 12 = 31 .
The same conditional probability also follows from the definition.

Multiplication Rule
Rule 7: P(A AND B) = P(A | B) · P(B) (follows from the definition)

23 / 90
Example 1

Suppose you draw a number at random from the set of whole numbers
{1, 2, 3, ... , 20}. Here, the sample space

S = {1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20}.

Let the event A denote that the number drawn is an even number and
the event B denote that it is greater than 13.
Find the following probabilities:
P(A), P(B)
P(A AND B), P(A OR B)
P(A | B), P(B | A)

24 / 90
Example 1
Observe that
A = {2, 4, 6, 8, 10, 12, 14, 16, 18, 20}
B = {14, 15, 16, 17, 18, 19, 20}
A OR B = {2, 4, 6, 8, 10, 12, 14, 15, 16, 17, 18, 19, 20}
A AND B = {14, 16, 18, 20}
Hence,
10 1 7
P(A) = = , P(B) =
20 2 20
13 4 1
P(A OR B) = , P(A AND B) = =
20 20 5
P(A AND B)
P(A | B) = = (1/5)/(7/20) = 4/7
P(B)
P(A AND B) 2
P(B | A) = = (1/5)/(1/2) = .
P(A) 5
25 / 90
Example 2

Suppose you roll a six–faced fair die. Find the probability that the roll
results in a multiple of 3 given that it is already known that an even
number has appeared.
Let A be the event that a multiple of 3 appears and B be the event that
a even number appears.
Here, the sample space S = {1, 2, 3, 4, 5, 6}, A = {3, 6}, B = {2, 4, 6},
so that A AND B = {6}.
Hence.
P(A AND B) 1
P(A | B) = = .
P(B) 3

26 / 90
Example 3

A student goes to the library. Let events B = the student checks out a
book and D = the student checks out a DVD.
Suppose that P(B) = 0.4, P(D) = 0.3 and P(D | B) = 0.5.
1 Find P(B AND D)
2 Fine P(B OR D)

Using the multiplication formula (Rule 7) we obtain

P(B AND D) = P(B) · P(D | B) = 0.4 × 0.5 = 0.2

Again, applying the union formula (Rule 2) we get

P(B OR D) = P(B) + P(D) − P(B AND D) = 0.4 + 0.3 − 0.2 = 0.5

27 / 90
Independence of Events
Heuristic Definition
We say that two events A and B are independent if ‘information’
about event A doesn’t provide any ‘information’ about event B.
In other words, two events A and B are said to be independent if
and only if the occurrence or the non–occurrence of A doesn’t
have any impact on the occurrence or the non–occurrence of B.

Formal Mathematical Definition

There are three equivalent formal definitions:
Two events A and B are independent if and only if they satisfy either
of the following conditions:
P(A ∩ B) = P(A)P(B)
P(A | B) = P(A)
P(B | A) = P(B)
28 / 90
Example 1
Suppose a fair coin is tossed twice. Here, S = {HH, HT, TH, TT}.
Let
A = ”occurrence of a head in the 1st toss” = {HH, HT}
B = ”occurrence of a tail in the 2nd toss” = {HT, TT}
The intersection of these two events is A AND B = {HT}.
Note that:
P(A) = 2/4 = 1/2
P(B) = 2/4 = 1/2
P(A AND B) = 1/4

Therefore
P(A AND B) = P(A)P(B).
Hence, A and B are independent events.
29 / 90
Example 2

The success probability of a particular knee surgery performed by a

surgeon is 0.95. Suppose he operates two patients in a day. Assume
that the surgeries are done independent of each other. What is the
probability of exactly one successful surgery in a day?
Let S and F respectively denote the ”success” and the ”failure” of a
given surgery. Here, P(S) = 0.95 and P(F) = 1 − P(S) = 0.05.
Here, the sample space S = {SS, SF, FS, FF}.
We need to find the probability P({SF, FS}).
Question: Are the elements of S equally likely? NO!

30 / 90
Example 2

Observe that because of the independence of the surgeries

P(SS) = P(S)P(S) = 0.95 × 0.95 = 0.9025

P(SF) = P(S)P(F) = 0.95 × 0.05 = 0.0475
P(FS) = P(F)P(S) = 0.05 × 0.95 = 0.0475
P(FF) = P(F)P(F) = 0.05 × 0.05 = 0.0025.

Hence, the elements of S are not equally likely.

Now, since the events {SF} and {FS} are disjoint, we have

P({SF , FS}) = P(SF ) + P(FS) = 2 × 0.0475 = 0.095.

Note: Since the elements of S are not equally likely, the probability of
exactly one successful surgery will not be 14 in this case.

31 / 90
Independence and Disjointedness are different

Don’t be confused between independent events and disjoint events!

If A and B with P(A) > 0 and P(B) > 0 are independent, they
cannot be disjoint (since in that case P(A AND B) > 0).

Equivalently, if A and B with P(A) > 0 and P(B) > 0 are disjoint,
they cannot be independent (since in that case Ac ⊂ B and
Bc ⊂ A).

However, if A and B with P(A) > 0 and P(B) > 0 are NOT disjoint,
there’s no guarantee they will be independent.

32 / 90
Independence and Disjointedness are different
Suppose a fair coin is tossed twice. Here, S = {HH, HT, TH, TT}.
Let
A = ”at least one head occurs” = {HH, HT, TH}
B = ”at least one tail occurs” = {HT, TH, TT}

The intersection of these two events is A AND B = {HT, TH} =

6 ∅.
Note that:
P(A) = 3/4
P(B) = 3/4
P(A AND B) = 2/4 = 1/2

Therefore
P(A AND B) 6= P(A)P(B).
Hence, A and B are not independent events, and are not disjoint also.
33 / 90
Independence and Disjointedness are different
Suppose a fair six-faced die is thrown once. Here, S = {1, 2, 3, 4, 5, 6}.
Let
A = ”an even number appears” = {2, 4, 6}
B = ”a multiple of 3 appears” = {3, 6}

The intersection of these two events is A AND B = {6} =

6 ∅.
Note that:
P(A) = 3/6 = 1/2
P(B) = 2/6 = 1/3
P(A AND B) = 1/6

Therefore
P(A AND B) = P(A)P(B).
Hence, A and B are independent events, and they are not disjoint.
34 / 90
Sampling Effect Probabilities

Sampling may be done with replacement or without replacement.

With replacement: If each member of a population is replaced after it
is picked, then that member has the possibility of being chosen more
than once. When sampling is done with replacement, then events are
considered to be independent, meaning the result of the first pick will
not change the probabilities for the second pick.
Without replacement: When sampling is done without replacement,
each member of a population may be chosen only once. In this case,
the probabilities for the second pick are affected by the result of the first
pick. The events are considered to be dependent or not independent.

35 / 90
Illustrative Example
You have a fair, well-shuffled deck of 52 cards. It consists of four suits.
The suits are clubs, diamonds, hearts and spades. There are 13 cards
in each suit consisting of 2, 3, 4, 5, 6, 7, 8, 9, 10, J (Jack), Q (Queen),
K (King) and A (Ace) of that suit.

Let’s look at what sampling with replacement and without replacement

each look like in this context.
36 / 90
Sampling with replacement

Suppose you pick three cards with replacement. You pick each card
from the 52-card deck.
1 Q of Spades (1 out of 52 cards)
2 10 of Clubs (1 out of 52 cards)
3 Q of Spades (1 out of 52 cards)
Sampling with replacement allows you to get the same card twice.

37 / 90
Sampling without replacement

Suppose you pick three cards without replacement.

1 K of Hearts (1 out of 52 cards)
2 3 of Diamonds (1 out of 51 cards)
3 J of Spades (1 out of 50 cards)
Because you have picked the cards without replacement, you cannot
pick the same card twice.

38 / 90
Tree Diagrams

A tree diagram is a special type of graph used to determine the

outcomes of an experiment. It consists of ”branches” that are labeled
with either frequencies or probabilities. Tree diagrams can make some
probability problems easier to visualize and solve.

39 / 90
Tree Diagrams

In an urn, there are 11 balls. Three balls are red (R) and eight balls are
blue (B). Draw two balls, one at a time, with replacement. The tree
diagram using frequencies that show all the possible outcomes follows.

40 / 90
Tree Diagrams

41 / 90
Tree Diagrams

a. List all 24 BR outcomes.

b. Using the tree diagram, calculate P(RR). Note RR is the event that
we draw two red balls.
c. Using the tree diagram, calculate P(RB OR BR).
d. Using the tree diagram, calculate P(R on 1st draw AND B on 2nd
draw).
e. Using the tree diagram, calculate P(R on 2nd draw GIVEN B on 1st
draw).
f. Using the tree diagram, calculate P(BB).
g. Using the tree diagram, calculate P(B on the 2nd draw given R on
the first draw).

42 / 90
Tree Diagrams

a. List all 24 BR outcomes.

Outcomes with blue 1 first B1R1; B1R2; B1R3;
Outcomes with blue 2 first B2R1; B2R2; B2R3;
Outcomes with blue 3 first B3R1; B3R2; B3R3;
Outcomes with blue 4 first B4R1; B4R2; B4R3;
Outcomes with blue 5 first B5R1; B5R2; B5R3;
Outcomes with blue 6 first B6R1; B6R2; B6R3;
Outcomes with blue 7 first B7R1; B7R2; B7R3;
Outcomes with blue 8 first B8R1; B8R2; B8R3;

43 / 90
Tree Diagrams
b. Using the tree diagram, calculate P(RR). Note RR is the event that
we draw two red balls.
To calculate this we multiply the probabilities down the tree. Since we
are drawing with replacement the probability of drawing a red is 3/11
each time, so
3 3 9
P(RR) = =
11 11 121

44 / 90
Tree Diagrams
c. Using the tree diagram, calculate P(RB OR BR).
They are disjoint since we cannot both draw a blue ball first and a red
ball first, so we add the probabilities events.

P(RB OR BR) =

45 / 90
Tree Diagrams

d. Using the tree diagram, calculate

P(R on 1st draw AND B on 2nd draw).

P(R on 1st draw AND B on 2nd draw) = P(RB) =

46 / 90
Tree Diagrams

e. Using the tree diagram, calculate

P(R on 2nd draw GIVEN B on 1st draw).
This problem is a conditional one. There are 24 + 64 = 88 possible
outcomes (24 BR and 64 BB).

P(R on 2nd draw GIVEN B on 1st draw) = P(R|B) =

Number of BR outcomes
=
total number outcomes where we drew a blue marble first

47 / 90
Tree Diagrams

f. Using the tree diagram, calculate P(BB).

P(BB) =

48 / 90
Tree Diagrams

g. Using the tree diagram, calculate

P(B on the 2nd draw given R on the first draw).

P(B|R) =

49 / 90
Without Replacement

We will now look at the problem again but this time we are drawing
without replacement. This will change the probability calculations.

50 / 90
Without Replacement

An urn has three red marbles and eight blue marbles in it. Draw two
marbles, one at a time, this time without replacement, from the urn.
The branches are labeled with probabilities instead of frequencies.
The numbers at the ends of the branches are calculated by multiplying
the numbers on the two corresponding branches, for example, the
probability of drawing RR is

P(RR) =

51 / 90
Without Replacement
Note: You draw without replacement, so that on the second draw there
are ten marbles left in the urn.

52 / 90
Without Replacement

a. P(RR)
b. P(RB OR BR)
c. P(R | B)
d. P(R on 1st AND B on 2nd)
e. P(BB)
f. P(B | R)

53 / 90
Without Replacement
a. P(RR)

P(RR) =

54 / 90
Without Replacement
b. P(RB OR BR)

P(RB OR BR) =

55 / 90
Without Replacement
c. P(R | B)

P(R | B) =
Here we are only looking at the number of ways to draw a red ball in
the event that we have drawn a blue ball first.

56 / 90
Without Replacement
d. P(R on 1st AND B on 2nd)

P(R on 1st AND B on 2nd) =

57 / 90
Without Replacement
e. P(BB)

P(BB) =

58 / 90
Without Replacement
f. P(B | R)

P(B|R) =

59 / 90
Another Tree diagram

Large-scale surveys indicate that 11% of the population smokes.

Medical researchers know that the probability that a smoker will get
lung cancer is 0.34. The probability that a person will get lung cancer if
the person does not smoke is 0.03. S=smoker; NS=not smoker; LC=
lung cancer; NLC=not lung cancer

P(S) =

60 / 90
Another Tree diagram

Large-scale surveys indicate that 11% of the population smokes.

Medical researchers know that the probability that a smoker will get
lung cancer is 0.34. The probability that a person will get lung cancer if
the person does not smoke is 0.03. S=smoker; NS=not smoker; LC=
lung cancer; NLC=not lung cancer

P(S) = .11

61 / 90
Another Tree diagram

Large-scale surveys indicate that 11% of the population smokes.

Medical researchers know that the probability that a smoker will get
lung cancer is 0.34. The probability that a person will get lung cancer if
the person does not smoke is 0.03. S=smoker; NS=not smoker; LC=
lung cancer; NLC=not lung cancer

P(LC|S) =

62 / 90
Another Tree diagram

Large-scale surveys indicate that 11% of the population smokes.

Medical researchers know that the probability that a smoker will get
lung cancer is 0.34. The probability that a person will get lung cancer if
the person does not smoke is 0.03. S=smoker; NS=not smoker; LC=
lung cancer; NLC=not lung cancer

P(LC|S) = .34

63 / 90
Another Tree diagram

Large-scale surveys indicate that 11% of the population smokes.

Medical researchers know that the probability that a smoker will get
lung cancer is 0.34. The probability that a person will get lung cancer if
the person does not smoke is 0.03. S=smoker; NS=not smoker; LC=
lung cancer; NLC=not lung cancer

P(LC|NS) =

64 / 90
Another Tree diagram

Large-scale surveys indicate that 11% of the population smokes.

Medical researchers know that the probability that a smoker will get
lung cancer is 0.34. The probability that a person will get lung cancer if
the person does not smoke is 0.03. S=smoker; NS=not smoker; LC=
lung cancer; NLC=not lung cancer

P(LC|NS) = 0.03

65 / 90
Another Tree diagram

Large-scale surveys indicate that 11% of the population smokes.

Medical researchers know that the probability that a smoker will get
lung cancer is 0.34. The probability that a person will get lung cancer if
the person does not smoke is 0.03. S=smoker; NS=not smoker; LC=
lung cancer; NLC=not lung cancer

P(S and LC) =

66 / 90
Another Tree diagram

Large-scale surveys indicate that 11% of the population smokes.

Medical researchers know that the probability that a smoker will get
lung cancer is 0.34. The probability that a person will get lung cancer if
the person does not smoke is 0.03. S=smoker; NS=not smoker; LC=
lung cancer; NLC=not lung cancer

P(S and LC) = P(LC|S)P(S) = .34(.11) = 0.0374

67 / 90
Calculating Probabilities from Contingency Tables

A contingency table provides a way of portraying data that can

facilitate calculating probabilities. The table helps in determining
conditional probabilities quite easily. The table displays sample values
in relation to two different variables that may be dependent or
contingent on one another. Later on, we will use contingency tables
again, but in another manner.

68 / 90
Calculating Probabilities from Contingency Tables

Suppose a study of speeding violations and drivers who use cell

phones produced the following fictional data:

69 / 90
Calculating Probabilities from Contingency Tables

Calculate the following probabilities using the table.

a. Find P(Driver is a cell phone user).
b. Find P(driver had no violation in the last year).
c. Find P(Driver had no violation in the last year AND was a cell phone
user).
d. Find P(Driver is a cell phone user OR driver had no violation in the
last year).
e. Find P(Driver is a cell phone user GIVEN driver had a violation in
the last year).
f. Find P(Driver had no violation last year GIVEN driver was not a cell
phone user)

70 / 90
a. Add the row ’Uses cell’
Number of cell phone users 305
=
Total number in study 755

71 / 90
b. Add the column ’no violations’
Number of no violations 685
=
Total number in study 755

72 / 90
 280
c. Read this right off the chart 755

73 / 90
d. To calculate P(C OR NV) we need to use rule 3

P(C OR NV ) = P(C) + P(NV ) − P(C AND NV )

305 685 280 710

+ − =
755 755 755 755

74 / 90
e. To calculate P(C | NV) we need to consider just the number of ’no
violations’ column
25
P(C | NV ) =
70

75 / 90
f. To P(NV | Cc ) we need to consider just the number of ’not cellphone’

405
P(NV | C c ) =
450

76 / 90
Additional Problems

77 / 90
Problem # 1

J and K are independent events. P(J|K) = 0.3. Find P(J).

Independence means P(J|K) = P(J) = 0.3.

78 / 90
Problem # 2

U and V are mutually exclusive events. P(U) = 0.26; P(V) = 0.37. Find:
a. P(U AND V) = 0
b. P(U|V) = 0
c. P(U OR V) = P(U)+P(V)=0.26+0.37

79 / 90
Problem # 3
After Rob Ford, the mayor of Toronto, announced his plans to cut
budget costs in late 2011, the Forum Research polled 1,046 people to
measure the mayor’s popularity. Everyone polled expressed either
approval or disapproval. These are the results their poll produced:
In early 2011, 60 % of the population approved of Mayor Ford’s
actions in office.
In mid-2011, 57 % of the population approved of his actions.
In late 2011, the percentage of popular approval was at 42 %.
a. What is the sample size for this study?
b. What proportion in the poll disapproved of Mayor Ford, according to the
results from late 2011?
c. How many people polled responded that they approved of Mayor Ford in
late 2011?
d. What is the probability that a person supported Mayor Ford, based on the
data collected in mid-2011?
e. What is the probability that a person supported Mayor Ford, based on the
data collected in early 2011?
80 / 90
Problem # 3

Answers:

a. What is the sample size for this study? 1046

b. What proportion in the poll disapproved of Mayor Ford, according to
the results from late 2011? 1-0.42
c. How many people polled responded that they approved of Mayor
Ford in late 2011? 0.42(1046)
d. What is the probability that a person supported Mayor Ford, based
on the data collected in mid-2011? 0.57
e. What is the probability that a person supported Mayor Ford, based
on the data collected in early 2011? 0.60

81 / 90
Problem # 4

Suppose that you have eight cards. Five are green and three are
yellow. The five green cards are numbered 1, 2, 3, 4, and 5. The three
yellow cards are numbered 1, 2, and 3. The cards are well shuffled.
You randomly draw one card.
G = card drawn is green
E = card drawn is even-numbered
a. List the sample space. b. P(G)
c. P(G|E)
d. P(G AND E)
e. P(G OR E)
f. Are G and E mutually exclusive? Justify your answer numerically.

82 / 90
Problem # 4

Suppose that you have eight cards. Five are green and three are
yellow. The five green cards are numbered 1, 2, 3, 4, and 5. The three
yellow cards are numbered 1, 2, and 3. The cards are well shuffled.
You randomly draw one card.
G = card drawn is green
E = card drawn is even-numbered
a. S = {Y1, Y2, Y3, G1, G2, G3, G4, G5}
b. P(G) = 5/8
c. P(G|E) = 2/3
d. P(G AND E) = 2/8
e. P(G OR E) = 5/8 + 3/8 − 2/8 = 6/8
f. Are G and E mutually exclusive? Justify your answer numerically.
No, P(G AND E)6=0

83 / 90
Problem # 5

Find P(hit was made by Babe Ruth).

Find P(hit was made by Ty Cobb|The hit was a Home Run).

84 / 90
Problem # 5

Find P(hit was made by Babe Ruth)=2873/12351

Find P(hit was made by Ty Cobb|The hit was a Home Run)= 114/1720

85 / 90
Problem # 5

Are ”the hit being made by Hank Aaron” and ”the hit being a double”
independent events?
Yes, because P(hit by Hank Aaron|hit is a double) = P(hit by Hank
Aaron)
No, because P(hit by Hank Aaron|hit is a double) 6= P(hit is a double)
No, because P(hit is by Hank Aaron|hit is a double) 6= P(hit by Hank
Aaron)
Yes, because P(hit is by Hank Aaron|hit is a double) = P(hit is a
double)

86 / 90
Problem # 5

Are ”the hit being made by Hank Aaron” and ”the hit being a double”
independent events?
Yes, because P(hit by Hank Aaron|hit is a double) = P(hit by Hank
Aaron)
No, because P(hit by Hank Aaron|hit is a double) 6= P(hit is a double)
No, because P(hit is by Hank Aaron|hit is a double) 6= P(hit by
Hank Aaron)
Yes, because P(hit is by Hank Aaron|hit is a double) = P(hit is a
double)

87 / 90
Problem # 6

The percent of licensed U.S. drivers (from a recent year) that are
female is 48.60. Of the females, 5.03% are age 19 and under; 81.36%
are age 20–64; 13.61% are age 65 or over. Of the licensed U.S. male
drivers, 5.04% are age 19 and under; 81.43% are age 20–64; 13.53%
are age 65 or over. Complete the following.
a. Construct a table or a tree diagram of the situation.
b. Find P(driver is female).
c. Find P(driver is age 65 or over|driver is female).
d. Find P(driver is age 65 or over AND female).
e. In words, explain the difference between the probabilities in part c
and part d.
f. Find P(driver is age 65 or over).
g. Are being age 65 or over and being female mutually exclusive
events? How do you know?

88 / 90
Problem # 6
The percent of licensed U.S. drivers (from a recent year) that are
female is 48.60. Of the females, 5.03% are age 19 and under; 81.36%
are age 20-64; 13.61% are age 65 or over. Of the licensed U.S. male
drivers, 5.04% are age 19 and under; 81.43% are age 20-64; 13.53%
are age 65 or over. Complete the following.
a. Construct a table or a tree diagram of the situation. On Board
b. Find P(driver is female)= 0.486.
c. Find P(driver is age 65 or over|driver is female)= 0.1361/0.486.
d. Find P(driver is age 65 or over AND female)= 0.1361.
e. In words, explain the difference between the probabilities in part c
and part d.
f. Find P(driver is age 65 or over)= 0.1361 + 0.1353.
g. Are being age 65 or over and being female mutually exclusive
events? How do you know? No P(driver is age 65 or over AND
female)6= 0
89 / 90
This OpenStax ancillary resource is copyright Reza Khademakbari
under a CC BY-NC-SA 4.0 International license; it may be reproduced
or modified for noncommercial purposes only but must be attributed to
Reza Kahdemakbari and any changes must be noted. Any adaptation
must be shared under the same type of license.

90 / 90