IV: DATA ANALYSIS AND

INTERPRETATION Angelo M. Alberto

PROBABILITY
 SOME STATEMENTS
INVOLVING PROBABILITIES
EXAMPLE 1
In an employee performance test, possible
results are excellent, very good, good, fair,
and poor. If the probability of “excellent” in
a certain company is 0.1, it means that…

 if a large number of employees are

subjected to such assessment tests, around
10% of them will be “excellent”; and/or

if one employee is selected randomly, there

is a 10% chance that s/he is “excellent”
EXAMPLE 2
The probability of developing breast
cancer over 30 years among 40 year old
women is approximately 1/11. This means
approximately 1 in 11 of these 40 year old
women will have developed breast cancer
by the age of 70.
(Of course, aside from age, there are other
factors explaining likelihood of getting breast
cancer)
EXAMPLE 3
The probability of obtaining similar numbers
when a pair of dice is thrown is 6/36.
EXAMPLE 4
When a coin is tossed three times, there is a
6/8 chance of obtaining dissimilar faces
INTRODUCTORY
TERMS
Statistical experiment
Sample space
Sample point
Event
STATISTICAL EXPERIMENT
A controlled repeatable procedure that
generates well-defined outcomes
Soon, generates DATA

For each run of the experiment, only one of all

possible outcomes occurs
1: Randomly selecting an employee in a
company and checking his/her performance
2: Randomly selecting a female patient and
checking if she has breast cancer
3: Throwing two fair dice or throwing a fair die
twice
4: Tossing three fair coins or tossing one fair
coin thrice
SAMPLE SPACE
The set of all possible outcomes of an
experiment
Usually denoted by “S”
Analogous to the universal set U (in
mathematics)
1: S = {Excellent, Very Good, Good, Fair,
Poor}
2: S = {No cancer, with cancer}
3: S = {11,12,13,….., 64, 65, 66}
4: S = {HHH, HHT, HTT, …. , TTT}
SAMPLE POINT
Another term for an experimental outcome
An element of the sample space
1: Five sample points/ experimental
outcomes: Excellent, Very Good, Good, Fair,
Poor
2: Two sample points: No cancer, with cancer
3: 36 sample points: 11,12,13,….., 64, 65, 66
4: Eight sample points: HHH, HHT, HTT, …. ,
TTT
One may construct a tree diagram to map all
possible outcomes of an experiment
EVENT
A collection of sample points
Any subset of the sample space
 May contain any of the following:
 No sample point
 One sample point (simple event or elementary
event)
 More than one sample point (compound or composite
event)
 All sample points in the sample space
Usually denoted by an uppercase letter
SOME EXAMPLES OF EVENTS
1: J= the event of not being excellent
J = {Very Good, Good, Fair, Poor}

3: K = the event of getting even numbers

K = {22, 24, 26, 42, 44, 46, 62, 64, 66}

4: L = the event of getting dissimilar faces

L = {HHT, HTH, THH, TTH, THT, HTT}
 Important notation:
The uppercase P, then open and close
parentheses

P( )
PROBABILITY OF AN EVENT
The probability of an event A is denoted by
P(A)
In general, the probability of an event is
denoted by P(the event)
EXTREME TYPES OF EVENTS
Sure event

Impossible/null/empty event
BASIC PROPERTIES OF
PROBABILITIES
1) The probability of an event A, denoted by
P(A), always satisfies 0 < P(A) < 1

2) The sum of the probabilities of all possible

outcomes of an experiment is 1
3) If events A and B cannot happen at the
same time, then P(A or B) = P(A) + P(B)

In general, if n events A1, A2, … , An cannot

happen at the same time…
 P (A1 or A2 or … or An ) = P(A1) + P(A2) + …
+ P(An )
FROM EXAMPLE 1
Which of the following possible P(poor) is
acceptable?
P(poor) = -0.1
P(poor) = 1.01
P(poor) = 0
FROM EXAMPLE 1
Which of the following possible P(poor) is
acceptable?
P(poor) = -0.5
P(poor) =
0.690347690347687425689408087
P(poor) = 1.3
FROM EXAMPLE 1
If the probability of each of the first four
ratings is as follows, what is the probability of
“poor”? Rating P(Rating)

Excellent 0.1
Very good 0.3

Good 0.3
Fair 0.2
Poor ???
FROM EXAMPLE 1
What is the probability that a randomly selected
employee from the company is good or fair?
P(good) + P(fair) = 0.3 + 0.2 = 0.5

…is at least good?

P(excellent) + P(very good) + P(good) = 0.1 +
0.3 + 0.3 = 0.7

… is at most good?
P(good) + P(fair) + P(poor) = 0.3 + 0.2 + 0.1 =
0.6
FROM EXAMPLE 3
What is the probability of getting a 1,1?
P(1,1) = 1/36

What is the probability of getting a 1,1 or a 2,3?

P(1,1) + P(2,3) = 1/36 + 1/36 = 2/36

What is the probability of getting both even

numbers?
P(both even) = P(2,2) + P(2,4) + P(2,6) + P(4,2)
+ P(4,4) + P(4,6) + P(6,2) + P(6,4) + P(6,6) =
9(1/36) = 9/36
RECALL: BASED ON COUNTING
QUESTIONS FROM PREVIOUS PPT

What is the probability of selecting 6 red balls

from a jar of 30 balls – with 15 red, 10 blue and
5 yellow?

What is the probability of selecting 3 red balls, 2

blue balls, and 1 yellow ball from a jar of 30
balls – with 15 red, 10 blue and 5 yellow?
RECALL:

What is the probability that if 4 male students

and 3 female students are randomly lined up
for a photo, the three end up beside each
other?

… do not end up beside each other?

Don’t forget, combinations and permutations
may be used to determine the numerator and
denominator of a probability
BASIC RELATIONSHIPS IN
PROBABILITY
(AND PROCEDURES
ASSOCIATED WITH SUCH
RELATIONSHIPS)
OVERVIEW
Complement Intersection
 Independent events
 With replacement
Union sampling
Mutually exclusive  Dependent events
 Without replacement
events sampling
Non mutually  Conditional probability
exclusive events  Multiplication Law
Addition Law
One may construct a Venn diagram to show
complements, unions, and intersections
EXAMPLE 7 (RECURRING)
S = {1, 2, 3, 4, 5, 6}

Let
X = the event of getting an even number
Y = the event of getting a number greater
than 4
Z = the event of getting a 1
COMPLEMENT
Given an event A, the complement of A is the
event having all sample points which are NOT
in A, but are still in S
•Denoted by A’ (read as A prime)
•P(A) + P(A’) = 1
•P(A) = 1 –P(A’)
FROM EXAMPLE 7
What is X’? Y’? Z’?

What is P(X’)? P(Y’)? P(Z’)?

FROM EXAMPLE 1
What is the probability of a randomly selected
employee not being excellent?
P(excellent’) = 1 – 0.1 = 0.9

Or, if we let E = excellent…

P(E) = 0.1
P(E’) = 0.9
FROM EXAMPLE 3
What is the probability of not getting both
even numbers?
P(both even’) = 1 – 9/36 = 27/36

Note that in this example, “not both even” is

the same as “at least one odd”
FROM EXAMPLE 4
What is the probability of tossing a coin thrice
and not getting similar faces?
P(similar faces) = P(HHH or TTT) = P(HHH) +
P(TTT) = 1/8 + 1/8 = 2/8
Thus, P(dissimilar faces) = P(similar faces’) =
1 – 2/8 = 6/8
EXAMPLE 8
A jar has 15 red balls, 10 blue balls, and 5
yellow balls. You are to draw one ball
randomly. What is the probability that this ball
is not yellow?
Let R = red, B = blue, Y = yellow
P(not Y) = P(Y’) = 1 – 5/30 = 25/30
EXAMPLE 9
A class has 15 CED students, 10 COB
students, and 5 CLA students. You are to draw
one student randomly. What is the probability
that this student is not from CLA?
Let E = educ, B = business, L = liberal arts
P(not L) = P(L’) = 1 – 5/30 = 25/30
UNION
The union of events A and B, denoted by AUB,
is the event where either A occurs, B occurs,
or they both occur
In general, denoted by A1UA2UA3U….UAj,
where j is the number of events of interest

Usual keywords: “Or,” “Either,” “Any”

FROM EXAMPLE 7
What is XUY? XUZ? YUZ?

What is P(XUY)? P(XUZ)? P(YUZ)?

FROM EXAMPLE 1
What is the probability that a randomly selected
employee from the company is good or fair?
Let G = good, F = fair
P(G or F) = P(GUF) = P(G) + P(F) = 0.3 + 0.2 =
0.5

… is at least good?
P(E or V or G) = P(E) + P(V) + P(G) = 0.1 + 0.3
+ 0.3 = 0.7
FROM EXAMPLE 8
A jar has 15 red balls, 10 blue balls, and 5
yellow balls. You are to draw one ball
randomly. What is the probability that this ball
is yellow or blue?
Let R = red, B = blue, Y = yellow
P(B or Y) = P(BUY) = P(B) + P(Y) = 10/30 +
5/30 = 15/30
FROM EXAMPLE 9
A class has 15 CED students, 10 COB
students, and 5 CLA students. You are to draw
one student randomly. What is the probability
that this student is from COB or CLA
Let E = educ, B = business, L = liberal arts
P(B or L) = P(BUL) = P(B) = P(L) = 10/30 +
5/30 = 15/30
Examples under probability property 3 may
be linked to union, through the concept of
mutually exclusive events
RECALL: PROBABILITY
PROPERTY 3
If events A and B cannot happen at the same
time, then P(A or B) = P(A) + P(B)

In general, if n events A1, A2, … , An cannot

happen at the same time…
 P (A1 or A2 or … or An ) = P(A1) + P(A2) + …
+ P(An )
 MUTUALLY EXCLUSIVE
EVENTS
Events which cannot occur at the same
time
“Or” may be replaced by U
RECALL: PROBABILITY
PROPERTY 3
If events A and B cannot happen at the same
time, then P(A U B) = P(A) + P(B)

In general, if n events A1, A2, … , An cannot

happen at the same time…
 P (A1 U A2 U … U An ) = P(A1) + P(A2) + … +
P(An )
 ADDITION LAW OF
PROBABILITIES
If events A and B cannot occur at the same
time, then P(AUB) = P(A) + P(B)

Applies also to multiple events

 ADDITION LAW OF
PROBABILITIES
But, if events A and B can occur at the same
time, then P(AUB) = P(A) + P(B) - P(A∩B)

what is ∩ ?
INTERSECTION
The intersection of events A and B, denoted
by A ∩ B, is the event where BOTH A and B
occur
In general, denoted by A1 ∩ A2 ∩ A3 ∩ …. ∩
Aj, where j is the number of events of interest

Usual keywords: “And,” “Both,” “All”

Mutually exclusive events do not have an
intersection. That is, for two mutually exclusive
events A and B, A ∩B = Ø and P(A ∩B) = 0

If events A and B cannot happen at the same

time, then P(A or B) = P(A) + P(B) - P(A ∩B)
Here, P(A ∩B) = 0

BUT if events A and B can happen at the same

time, then P(AUB) = P(A) + P(B) - P(A∩B)
Because P(A ∩B) here is NOT ZERO
FROM EXAMPLE 7
What is X ∩ Y? X ∩ Z? Y ∩ Z?

What is P(X ∩ Y)? P(X ∩ Z)? P(Y ∩ Z)?

FROM EXAMPLE 1
What is the probability that a randomly
selected employee from the company is good
AND fair?
Let G = good, F = fair
G ∩F = Ø
P(G and F) = P(G ∩ F) = 0
FROM EXAMPLE 8
A jar has 15 red balls, 10 blue balls, and 5
yellow balls. You are to draw one ball
randomly. What is the probability that this ball
is yellow AND blue?
Let R = red, B = blue, Y = yellow
P(B and Y) = P(B ∩ Y) = 0
EXAMPLE 10
A study of streams near an industrial complex
revealed that 35% of them have a high BOD,
10% have high acidity, and 4% have both
characteristics.
If we are to randomly select one stream, and
test its water, what is the probability that this
particular stream has...

a)High acidity but does not have high BOD?

b)Neither high BOD nor high acidity?

Union is to mutual exclusivity as intersection
is to INDEPENDENCE
INDEPENDENT EVENTS
Intuitively, A is independent of B if the
probability of A occurring is not influenced or
explained by B, and vice versa

Two events A and B are independent if

P(A ∩B) = P(A)P(B)
This is the multiplication law for
independent events
NON-INDEPENDENT EVENTS
Intuitively, A is NOT independent of B
if the probability of A occurring is
influenced or explained by B

Two events A and B are not

independent if P(A ∩B) ≠ P(A)P(B)
Hypertension of mother and her child

Vs.

Hypertension of mother and father of the

child
EXAMPLE 11
What is the probability that both mother and
father are hypertensive if the probability that
mother is hypertensive is 0.1, and that of
father is 0.2?
EXAMPLE 12
What is the probability that both mother and
her child are hypertensive if the probability
that mother is hypertensive is 0.1, and that
of child is 0.05?
EXAMPLE 13
A container has 8 white pins and 12 black
pins. We are to draw two pins.

After the first pin is drawn, it is returned to

the container, and a second pin is drawn.
What is the probability that ...
-the two pins are white?
-the first pin is white and the second black?
EXAMPLE 14
A container has 8 white pins and 12 black
pins. We are to draw two pins.

After the first pin is drawn, it is NOT returned

to the container, and a second pin is drawn.
What is the probability that ...
-the two pins are white?
-the first pin is white and the second black?
Example 13 involves WITH REPLACEMENT
sampling

Example 14 involves WITHOUT REPLACEMENT

sampling
 SAMPLING WITH
REPLACEMENT
Selection of n units from a pool of N units
and placement of every selected unit back
to the pool of options

At draw 1, there are N options

At draw 2, there are N options
At draw 3, there are N options
And so on…

Usually involves principles of INdependence

 SAMPLING WITHOUT
REPLACEMENT
Selection of n units from a pool of N units
without returning the selected units to the
pool
At draw 1, there are N options
At draw 2, there are N-1 options
At draw 3, there are N-2 options
And so on…

Usually involves principles of dependence

If there’s no mention of which method to use –
that is, whether to use with replacement or
without replacement – assume use of without
replacement
Compilation of some keywords
Intersection of events usually involves
multiplication
and
union of events usually involves
addition
USUALLY, IN SOLVING…
Add probabilities if… Multiply probabilities if…
• Union • Intersection
• Or, any, either • And, both, all,
• Mutually exclusive • Independent /
/ non mutually dependent
exclusive
USUALLY, not ALL THE TIME
EXAMPLE 14
From previous experience of a certain laboratory,
the probability that the test tube is in good
condition is 0.95, and the probability that the
heater is in good condition is 0.80. At this
particular laboratory, what is the probability that...

•both the test tube and the heater are in good

condition?

•at least one of them is in good condition?

RECALL: DEPENDENT
EVENTS
Intuitively, A is not independent of B
if the probability of A occurring is
influenced or explained by B

Two events A and B are not

independent if P(A ∩B) ≠ P(A)P(B)
So what is P(A ∩B) equal to?
 P(A ∩B) = P(A)P(B l A)

P(B l A) = the probability of B occurring, given that A

has already occurred
- CONDITIONAL PROBABILITY 
CONDITIONAL PROBABILITY
If A and B are any two events in S, then the conditional
probability of B given A is

P(A ∩B) = P(A)P(B l A)

P(A ∩B) / P(A) = P(B l A)

P(B l A) = P(A ∩B) / P(A)

Note, P(B l A) is not necessarily equal to
P(A l B)
EXAMPLE 16
Let A be the event that the number revealed
by a die is odd, and B be the event that the
number is less than 3. What is P(B l A)?
EXAMPLE 17
Powdered Chocolate Liquid
Chocolate Syrup Chocolate Total
Pre-school 42 21 45 108
Grade 98 79 91 268
school
High 64 56 53 173
school
Total 204 156 189 549
What is the probability that a student likes syrup given that
s/he is in high school?
What is the probability that a student is in grade school
given that s/he likes liquid?
OTHER EXAMPLES
EXAMPLE 18
Jar A has 3 yellow balls and 2 red balls
Jar B has 7 yellow balls and 4 red balls
Randomly draw one ball from each jar

What is the probability that these two balls are both yellow?
What is the probability that these two balls are of the same
color?
What is the probability that these two balls are of different
colors?
EXAMPLE 19
Jar A has 3 yellow balls and 2 red balls
Jar B has 7 yellow balls and 4 red balls
Randomly draw one ball from Jar A, transfer the
ball to Jar B, then randomly draw one ball from
Jar B

What is the probability that the ball drawn from

Jar B is yellow?
EXAMPLE 20: FROM
EXAMPLE 8
A jar has 15 red balls, 10 blue balls, and 5
yellow balls. You are to draw 6 balls randomly.
What is the probability that of the 6 balls
drawn...
All are red?
5 are yellow?
3 are red, 2 are blue, and 1 is yellow?
EXAMPLE 21: FROM
EXAMPLE 9
A class has 15 CED students, 10 COB
students, and 5 CLA students. You are to draw
6 students randomly. What is the probability
that of the 6 students drawn...
All are from CED?
5 are from CLA?
3 are from CED, 2 are from COB, and 1 is
from CLA?
EXAMPLE 22
In a survey on students’ chocolate
preference, 8 must be sampled from a class
of 41. Assuming that there are 12 low-
income, 22 middle-income, and 7 high-
income students in the class, what is the
probability that the sample will…
Consist of all high-income?
Consist of all low-income?
Resemble that from stratified-by-income
sampling?
MORE EXAMPLES in the book
STEP 4: DATA ANALYSIS AND
INTERPRETATION Angelo M. Alberto

PROBABILITY