TEST OF HYPOTHESIS

By Engr. John Joseph C. Fernando

The main purpose of Statistics is to
answer questions about the world. It
makes assumptions believable.
Simply, it takes those assumptions,
gather information about that, test
those data, and draw conclusions.
Test of Hypothesis
Is a formal procedure that allows you to test
assumptions or so called HYPOTHESES
regarding a population parameter.

It assess the plausibility of a hypothesis by

using sample data.
Steps in Test of Hypothesis
1.) Formation of Hypotheses
2.) Formulation of Analysis Plan
3.) Analysis of Sample Data
4.) Interpretation of Results
STEP 1: Formation of Hypotheses
Hypotheses are the assumptions or
educated guesses made by the statisticians
or researcher about something.

There are two hypotheses being made in

every situation being tested.
STEP 1: Formation of Hypotheses
In every situation, you will always have to come up
with your NULL HYPOTHESIS (H0) and ALTERNATIVE
HYPOTHESIS (H1).
Example:
Situation: Michelle heard from the news that the solar system will
now only have 8 planets. She insists that Pluto is still a planet.
H0: Pluto is a planet.
H1: Pluto is not a planet.
LET’S TRY
Chinese researchers have come up with a medicine for corona
virus. They claim it is effective.

Abby decides to sell Chocolate chip cookies and thinks that it

will be loved by employees in Makati area.

Pia assumes that the number of stray dogs in Malolos is two

thousand.
STEP 1: Formation of Hypotheses
EXAMPLE: For Mean
a.) The Jollibee manager in Malolos claims that the daily
amount of customers they have is 3,000.

b.) The Jollibee manager in Malolos claims that the daily

amount of customers they have is at least 3,000.

c.) The Jollibee manager in Malolos claims that the daily

amount of customers they have is at most 3,000.
STEP 1: Formation of Hypotheses
EXAMPLE: For Proportion
a.) The McDonalds manager in Malolos claims that the daily
amount of customers they have is 80% of the citizens.

b.) The McDonalds manager in Malolos claims that the daily

amount of customers they have is at least 80% of the citizens.

c.) The McDonalds manager in Malolos claims that the daily

amount of customers they have is at most 80% of the citizens.
 EXTRA NOTE
Ho always has a symbol with an equal in it.
H1 never has a symbol with an equal in it.
The choice of symbol depends on the wording of
the hypothesis test.

However, be aware that many researchers use = in the null

hypothesis, even with > or < as the symbol in the alternative
hypothesis. This practice is acceptable because we only make
the decision to reject or not reject the null hypothesis.
STEP 1: Formation of Hypotheses
Note that statisticians and researchers actually try
to understand real life situations to
acquire their null and alternative hypotheses. They
do not just solve problems on paper but actually
see the world and translate it to equations.
This kind of attitude from these people actually
save lives, earn more money, connect to others
more, etc.
STEP 2: Formulation of Analysis Plan
You now come up with your hypotheses. What now?
What will you do with it?

Analysis Plan is your understanding what kind of test

and analysis will you subject your hypotheses with.

The Analysis Plan specifies two things:

■ Significance Level
■ Test Method
STEP 2: Formulation of Analysis Plan
SIGNIFICANCE LEVEL
This the probability of your null hypothesis being rejected
when it is TRUE.
Commonly, researchers use the significance level of 0.05

However, some field like the Health and Medicine will

consider a much lesser significance level like 1% for their
researches.
 STEP 2: Formulation of Analysis Plan
Test Method
SITUATION TEST METHOD

For testing population proportions. If null Two-tail Z-test

hypothesis is “=“ and the alternative hypothesis
is “≠”
For testing population proportions. If null One-tail Z-test
hypothesis is either “>=“ or “<=“
For testing population mean. If null hypothesis Two-tail t-test
is “=“ and the alternative hypothesis is “≠”
For testing population mean. If null hypothesis One-tail t-test
is either “>=“ or “<=“
 STEP 3: Analysis of Sample Data
Based on situations or what the statisticians want to test,
there are several methods to test the hypotheses.

However, commonly we only have to find three values to

analyze the sample data:
■ Standard Error (for mean) / Standard Deviation (for
proportion)
■ Test Statistic (t-score or Z-score)
■ P-value
STEP 3: Analysis of Sample Data
For Proportion
𝑝(1−𝑝)
Standard Deviation σ=
𝑛

ො
(𝑝−𝑝)
Z-score z=
𝜎

P-value is the probability of the z-score that can be seen in

the Z-table.
STEP 3: Analysis of Sample Data
For Mean
𝑠
Standard error 𝑆𝐸 =
𝑛

Degree of freedom 𝐷𝐹 = 𝑛 − 1

(𝑥ഥ − 𝜇)
t-score 𝑡=
𝑆𝐸

P-value is the probability of the t-score that can be seen in

the t-table.
Remember
Situation What to do for P-value
If H1 is “less than” in Proportion The probability you have seen in
Testing the Z-table will be its P-value
If H1 is “greater than” in ProportionThe probability you have seen in
Testing the Z-table will be deducted to 1
If H1 is “less than” in Mean Testing The probability you have seen in
the t-table will be its P-value
If H1 is “greater than” in Mean The probability you have seen in
Testing the t-table will be deducted to 1
If H1 is “not equal” in Mean Testing The probability you have seen in
or Proportion Testing the t-table or z-table will be
multiplied to 2
STEP 4: Interpretation of Result
The results of test of hypothesis is fairly easy:

■ If your P-value is less than the Significance Level (α), we

will NOT ACCEPT the null hypothesis.
■ If your P-value is greater than the Significance level (α),
we will ACCEPT the null hypothesis.
■ If your P-value is equal to the Significance Level (α), it is
the decision of the analyst if he/she will accept or not
the null hypothesis. However, most researchers will
consider this case for NOT ACCEPTANCE.
For Testing Hypotheses about Population Proportion
Example
The Jollibee manager in Malolos claims that 75% of
the citizens likes their food. The McDonalds manager
cannot believe this and so he surveyed 50 people to
choose between Jollibee and McDonalds. He found
that 30 among them likes Jollibee. Assuming that the
people are normally distributed, test if the manager of
Jollibee is right with a significance level of 10%.
For Testing Hypotheses about Population Proportion
Example
An alien empire is considering taking over planet Earth,
but they will only do so if the portion of rebellious
humans is less than 10 percent. They abducted a
random sample of 100 humans, performed special
psychological tests, and found that 20 of the sample
are rebellious. Come up with the hypotheses for the
aliens and test them using 1% significance level.
For Testing Hypotheses about Population Proportion

Example
According from the news, 65% of the newborn babies
this year are females. You asked your local hospital
about this and they said that from 35 that was given
birth in their facility, only 10 are females. Assuming the
babies are normally distributed and using a 1%
significance level, test the news.
For Testing Hypotheses about Population Mean
Example
An inventor has developed a new, energy-efficient lawn
mower engine. He claims that the engine will run
continuously for 300 minutes on a single gallon of regular
gasoline. From his stock of 2000 engines, the inventor
selects a simple random sample of 30 engines for testing.
The selected engines run for an average of 295 minutes,
with a standard deviation of 20 minutes.
Come up with hypotheses and test them using a 0.05 level
of significance. (Assume that run times for the population of
engines are normally distributed.)
For Testing Hypotheses about Population Mean
Example
You heard your aunt that the average student count
per section in the Philippines is at most 60. You want
to verify this by counting students from 10 sections in
your school. From those sections in your school, you
calculated that the average student count is 45 with a
standard deviation of 10 students. Assuming that
those sections are normally distributed, test the
hypothesis you heard with a significance level of 1%.
For Testing Hypotheses about Population Mean
Example
Bon Air Elementary School has 1000 students. The
principal of the school thinks that the average IQ of
students at Bon Air is at least 110. To prove her point, she
administers an IQ test to 20 randomly selected students.
Among the sampled students, the average IQ is 108 with a
standard deviation of 10. Based on these results, should
the principal accept or reject her original hypothesis?
Assume a significance level of 0.01. (Assume that test
scores in the population of engines are normally
distributed.)