THE HYPOTHESIS

TESTING
Statistics and Probability
 LESSON OBJECTIVES

At the end of the session, you will be

able to:
● Define Hypothesis Testing
● Identify the Steps in Performing
Hypothesis Testing
● Perform the Hypothesis Testing
Procedure
 Population
At the onset of COVID-19, health
professionals are seeking for solution to
put an end to the pandemic – a long-term
solution to name. After a series of research
being conducted, they were able to find
the long-term solution they’ve been
aiming for – development of vaccine to
put an end to it.
Sample
This is where hypothesis tests come in. Its
role is to oversee population parameters
and sample statistics and verify if a
certain test done to one group of
samples is applicable for all. In this case,
whether the vaccine they were able to
develop is effective or not.
What is Hypothesis
Testing?

Why is it important and how it could

help me reach a decision?
 Let’s say we are inside a courtroom
attending a trial and the judge is sitting in
front
Hypothesis testing is very similar as to proving someone to
be guilty of something. Assuming you are accused of
cheating in a test. As established, you are innocent until
proven guilty.

The thought of you being innocent is our pre-

established judgment and the accusation of you
cheating is our attempt to contest the previous
statement, it is a claim that needs to be proven by the
people who accused you which is done in a logical
manner backed by evidences.
WHAT IS HYPOTHESIS
TESTING?

Hypothesis testing refers to the process of

making inferences or educated guesses about a
particular parameter. The procedure in summary
is pre-established statements vs new contesting
statements.
 Population
Example:
Pre-established Statement
A plant manager believes
that: mean fill of the
HYPOTHESIS
population of the Coca Cola
cans in 330ml.
Data sample collected
from the population
Does a sample mean of
New Contesting Statement
327ml provide sufficient
evidence to justify that mean The sample mean is
fill of the population is
327ml. Sample
330ml?

Reject or Do Not
Reject?
Hypothesis
 HOW TO TEST A HYPOTHESIS
To test a statistical hypothesis, one can perform the
following procedure.

01 02 03
State the null and Select a level of Select the
alternative significance appropriate test
hypotheses statistic

04 05 06
Formulate the Compute the Make a decision
decision rule value of the test
statistic
 01
State the Null or Alternative
Hypothesis
 Null hypothesis, denoted by 𝑯∅ or H null, is a
statement saying that there is no significant
difference between the population mean and the
value that is being claimed ( 𝜇 = 𝜇1 ) . It is the
hypothesis to be tested.
 Alternative hypothesis denoted by 𝑯𝒂 , is
a statement saying that there is a significant
difference between the population mean
and the value that is being claimed. This
hypothesis contains either a “<”, “>”, or “≠”.
 NULL HYPOTHESIS VS. ALTERNATIVE
HYPOTHESIS

NULL ALTERNATIVE
In a criminal trial, the null In the case that 𝑯𝟎 is rejected,
hypothesis is the statement the alternative hypothesis will be
saying that the defendant is true. Thus, the decision that “the
not guilty. defendant is guilty” follows.
QUICK TIP!
In stating the null hypothesis, look for the population mean or
proportion in the problem and use an equal sign =
In stating the alternative hypothesis, look for words or phrases
such as “is different”, “is greater than”, “is less than”, “increase”,
or “decrease”.
If the phrase is “is different”, use the inequality symbol ≠
If the phrase is “is greater than” or “is more than”, or “increase”
use the greater than symbol >
If the phrase is “is less than” or “decrease” use the less than
symbol <
Note: Most likely if the problem states more in
percentage, use p and if otherwise, use μ.
Example 1
The manager of a TV station claims that the mean
number of people watching their new sitcom in Manila
is 500,000 each day. A student researcher who wants
statistical evidence on this claim conducted her own
survey. Her survey resulted to a mean of 499,995.
Using a 0.05 level of significance, can it be concluded
that the mean number of people watching the new
sitcom is less than 500,000?
Null Hypothesis (𝑯∅ ) :
The mean number of people watching the new sitcom
in Manila is 500,000 each day 𝜇 = 500,000
Alternative Hypothesis (𝑯𝒂 ) :
The mean number of people watching the new sitcom
in Manila is less than 500,000 each day 𝜇 < 500,000
Example 2
The LB Company produces an average of 4%
defective bulbs each day. Recently, some of the
machines were upgraded. Out of 400 bulbs randomly
selected, only 18 bulbs were found defective. Does
this suggest that there is an increase in the number of
defective bulbs produced each day?

Null Hypothesis (𝑯∅ ) :

An average of 4% defective bulbs are produced each
day 𝜌 = 0.04
Alternative Hypothesis (𝑯𝒂 ) :
More than 4% defective bulbs are produced each day
𝜌 > 0.04
 LET’S TRY!
1. The Head of PE Department of a certain high school claims
that the mean height of Grade 7 students is 163 cm. The
mean height of 45 randomly selected Grade 7 is different
from 163 cm as claimed by the Head of the PE Department.
Formulate the null and alternative hypotheses.

2. It is claimed that 15% of the high school students are in

favor of online class. Of the 1,000 high school students
asked, only 14% are in favor of online class. Does this mean
that the percentage of high school students who are in favor
of online class is less than 15%? Formulate the null and
alternative hypotheses.
 02
Select a level of Significance
 Level of significance, alpha 𝛂, corresponds to the
area of the curve occupied by the rejection region. The
most used values for alpha are 0.01, 0.05 and 0.10.

If the level of significance used is 0.05 or 5%, then there

is a 5% chance that the null hypothesis would be rejected
when it should not be. That is, whenever the null hypothesis
is true, the researcher is 95% confident that he/she would
make the correct decision.
Level of Significance
• Is a probability of wrongly rejecting 𝐻∅ (when 𝐻∅ is
in fact true)
• It refers to how much error we allow to wrongly
reject 𝐻∅ .
Note: Lower significant level indicate that
you require stronger evidence before you
will reject the null hypothesis.

• 𝑎 = 0.05 (default value/consumer

research projects)
• 𝑎 = 0.01 (quality assurance)
• 𝑎 = 0.10 (political polling)
Level of Significance
 03
Select the appropriate Test Statistic
Test statistic is a standardized value computed in
hypothesis testing which becomes the basis to decide and
draw a conclusion.
Note: In a criminal trial, the test statistic is the evidence produced
by the prosecutor.

In this module, you will learn the two most used test statistic,
z-test and t-test.
 Z-Test
A z-test is a statistical test used for large sample, where the
sample size n is greater than or equal to 30 (𝑛 ≥ 30) and the
population standard deviation 𝜎 is known.
𝑥ҧ − 𝜇 𝑥ҧ − 𝜇
𝑧= 𝜎 𝑧= 𝑠
𝑜𝑟
𝑛 𝑛
Population variance is known Population variance is unknown
where: 𝒙
ഥ is the sample mean
𝝁 is the population mean
𝝈 is the population standard deviation
𝒔 is the sample standard deviation
𝒏 is the sample size
 T-Test
A t-test, on the other hand, is a statistical test used for
small sample, where the sample size n is less than 30 (𝑛 <
30) and the population standard deviation 𝜎 is unknown.
𝑥ҧ − 𝜇
𝑡= 𝑠
𝑛
where: 𝒙
ഥ is the sample mean
𝝁 is the population mean
𝝈 is the population standard deviation
𝒔 is the sample standard deviation
𝒏 is the sample size
 04
Formulate the decision rule
 The decision rule indicates the condition or
conditions where the null hypothesis is rejected. It
depends on whether the alternative hypothesis is
two-sided (with inequality symbol ≠) or follows a
two-tailed test, or one-sided (with symbol < or >)
or one-tailed test.
 Part of formulating the
decision rule is identifying the
The critical value is a particular point in the
critical value and the rejection rejection region that separates it from the
region. rest of the values. The critical value depends
on the level of significance of the test and
the distribution of the test statistic.
The rejection
region is the set
of values of the
test statistic
that result in the
rejection of the
null hypothesis.
 05
Compute the value of the Test
Statistic
 In computing the value of the test statistic, you
will identify the given values in the problem and
substitute those numbers to the appropriate test
statistic, and compute.
 06
Make a decision
 If the value of the test statistic falls within
the rejection region, reject 𝑯𝟎 in favor of the
alternative hypothesis. Otherwise, do not reject 𝑯𝟎 .

Not rejecting 𝑯𝟎 does not mean that 𝑯𝟎 is

true. It is just that, there is no sufficient evidence
or not enough evidence to say that it is false.
To easily interpret the decision, you can
use the following statement.

Claim: 𝑯𝟎 Claim: 𝑯𝒂

Reject 𝐻0 : There is Reject 𝐻0 : There is

enough evidence enough evidence
to reject the to support the
claim... claim...

Do not reject 𝐻0 : Do not reject 𝐻0 :

There is not There is not
enough evidence enough evidence
to reject the to support the
claim... claim..