The Language of Hypothesis Testing

Learning Objectives
❖Illustrate
–Null hypothesis
–Alternative hypothesis
–Level of significance
Learning Objectives
❖Illustrate
–Rejection region
–Types of error in hypothesis testing
Learning Objectives
❖Differentiate directional test from
nondirectional test
❖State and discuss the steps in hypothesis testing
❖Calculate the possibilities of committing type I
and type II errors
Inferential Statistics and Hypothesis Testing
The heart of statistics is the creation of
inferences or meaningful generalizations about a
given set of data or population.
Valid inferences can be made using only the
observations from the samples.
Hypothesis Testing
The process of making generalizations about
the characteristics of the entire population
through sample statistics.
Hypothesis
A tentative presupposition or an inference
made in order to predict the occurrence of a
phenomenon
Statistical Hypothesis
A claim about the value of a population
parameter or about the values of several
population parameters
Hypothesis Testing
1. State the hypothesis to be tested
2. Set the standard that describes whether the
claim is true or not
3. Compute the test statistic
4. Make the decision
Null Hypothesis
The statement of the value to which the
population parameter is equal and which is
presumed to be true.
It is being denoted by 𝐻𝑜 . A statement that
there is no difference between a parameter and a
specific value.
Alternative Hypothesis
It is the assertion that contradicts the null
hypothesis.
It is being denoted by 𝐻𝑎 .
It describes that the population parameter differs
from the population parameter predicted in the null
hypothesis.
Example 1
From the claim that
“The mean number of hours of sleep of SHS
students is 5”
Example
𝐻𝑜 : The mean number of hours of sleep of senior
high students is equal to 5.

𝐻𝑎 : The mean of the number of hours of sleep of

senior high students is not equal to 5.
Example 2
Tell whether the ff. statement is a null or an
alternative hypothesis: “The mean general
weighted average (GWA) of college students in
OLPCC is 84.8.”
Solution
“The mean general weighted average (GWA) of
college students in OLPCC is 84.8.”

Null Hypothesis
Solution
𝐻𝑜 :The mean general weighted average (GWA) of
college students in OLPCC is 84.8.
𝐻𝑎 : The mean GWA of college students in OLPCC is
not equal to 84.8
Solution
𝐻𝑜 :The mean general weighted average (GWA) of
college students in OLPCC is 84.8.
𝐻𝑎 :The mean GWA of college students in OLPCC is
greater than 84.8
Solution
𝐻𝑜 :The mean general weighted average (GWA) of
college students in OLPCC is 84.8.
𝐻𝑎 :The mean GWA of college students in OLPCC is
less than 84.8
Nondirectional Test
The design of the hypothesis test can either
be one – tailed or two – tailed.
The two – tailed test a.k.a. NONDIRECTIONAL
TEST is the standard test used in many researches
and it compares the population parameter in
both directions of the normal curve.
Directional Test
It is called the one – tailed test and it is a test
that determines the relationship between the
variables in only one direction, either the left or
the right tail of the normal curve.
Example
The alternative hypothesis “The mean number of
hours of sleep of SHS students is not equal to 5”

Non-directional hypothesis test

Example
Directional Test Form
❖The mean number of hours of sleep of SHS
students is greater than 5 (right – tailed)
❖The mean number of hours of sleep of SHS
students is less than 5 (left – tailed)
Directional vs Nondirectional
One – tailed Two – tailed
1. 𝐻𝑜 : 𝜇𝑜 = 𝜇 𝐻𝑜 : 𝜇𝑜 = 𝜇
𝐻𝑎 : 𝜇𝑜 < 𝜇 (left – tailed) 𝐻𝑎 : 𝜇𝑜 ≠ 𝜇
𝐻𝑎 : 𝜇𝑜 > 𝜇 (right – tailed)
2. 𝐻𝑜 : 𝜇𝑜 ≥ 𝜇
𝐻𝑎 : 𝜇𝑜 < 𝜇 (left – tailed)
3. 𝐻𝑜 : 𝜇𝑜 ≤ 𝜇
𝐻𝑎 : 𝜇𝑜 > 𝜇 (right – tailed)
Example
Given the null hypothesis, “The mean age of the
patients of Cauayan District Hospital is equal to 26”,
state the alternative hypothesis if the test is
a. Right – tailed
b. Left – tailed
c. Two – tailed
Solution (Right – tailed )
𝐻𝑜 : The mean age of the patients of CDH is equal
to 26 (𝐻𝑜 : 𝜇𝑜 = 26)
𝐻𝑎 : 𝜇𝑜 > 26
The mean age of the patients of CDH is greater
than 26.
Solution (Left – tailed )
𝐻𝑜 : The mean age of the patients of CDH is equal
to 26 (𝐻𝑜 : 𝜇𝑜 = 26)
𝐻𝑎 : 𝜇𝑜 < 26
The mean age of the patients of CDH is less than
26.
Solution (Two – tailed )
𝐻𝑜 : The mean age of the patients of CDH is equal
to 26 (𝐻𝑜 : 𝜇𝑜 = 26)
𝐻𝑎 : 𝜇𝑜 ≠ 26
The mean age of the patients of CDH is not equal
to 26.
The level of Significance
The decision to reject or to fail to reject the null
hypothesis is called significance and it should be
based on a set criterion of judgment called the
level of significance denoted by 𝛼.
Commonly used value of 𝛼 includes 1%, 5% and
10%.
The level of Significance
Significance is reached when the p – value
(probability value) of obtaining the sample
statistic is less than the set level of significance.
𝑡𝑒𝑠𝑡 𝑣𝑎𝑙𝑢𝑒 < 𝑐𝑟𝑖𝑡𝑖𝑐𝑎𝑙 𝑣𝑎𝑙𝑢𝑒
It is also the probability of committing type I
error.
Example

Decision:
Do not
reject Ho
Example

Decision:
Do not
reject Ho
Example

Decision:
Reject Ho
Example

Decision:
Do not
reject Ho
Example

Decision:
Reject Ho
Example

Decision:
Do not
reject Ho
Decision Errors in Hypothesis Testing
Generally, the decision errors committed during
hypothesis testing can be classified into two:
➢Type I: The null hypothesis is rejected when it is
true
➢Type II: The null hypothesis is not rejected when it
is false.
Decision Errors
Null Hypothesis
Do not Reject Reject
Value
TRUE Correct Type I
Decision error
FALSE Type II Correct
error Decision
Example (Type II error)
It has been established that a particular teaching
strategy improves math performance. However, the p
– value taken from your experiment at an alpha –
value of 0.05 was 0.15. Thus, you did not reject the
null hypothesis and concluded that there is no
significant difference between the teaching strategy
and math performance.
Activity A
Identify whether the following is a null hypothesis or
an alternative hypothesis
1. The average age of grade 11 students is 17 years
old.
2. The mean content of citric acid in a bottle of juice
drinks is greater than 2 ml.
Activity A
3. The average monthly salary of private school
teachers is less than P15, 000.
4. The mean weight of new born babies is 0.5 kg.
5. The average IQ of grade 11 students is less than
108.
Activity B
For each of the following conjecture, state the null
(Ho) and the alternative (Ha) hypotheses.
1. The average time to commute from home to
school is 30 minutes.
2. The average typing speed of a secretary is 25
words per minute.
Activity B
3. The average number of vehicles passing through
NLEX daily is less than 21,000.
4. The average daily number of customers in a
convenience store is less than 1, 000.
5. The mean content of sugar in a bottle of soda is
greater than 5 g.
Activity C
Using the hypotheses in Part B, identify whether
the test of hypothesis to be performed is one –
tailed or two – tailed.
Activity D
Identify whether the test is directional or non –
directional, given the ff.
1. 𝐻𝑜 : 𝜇 = 24
𝐻𝑎 : 𝜇 ≠ 24
Activity D
2. 𝐻𝑜 : 𝜇 = 24
𝐻𝑎 : 𝜇 < 24
3. 𝐻𝑜 : 𝜇 = 24
𝐻𝑎 : 𝜇 > 24
Activity E
Identify the type of Error if there is any committed in
each of the following situations.
1. A true null hypothesis is accepted
2. A false null hypothesis is accepted
3. A true null hypothesis is rejected
4. A false null hypothesis is rejected
Activity F
Identify whether the ff. statements are true or
false.
1. If the rejection region lies entirely on the left
tail of the sampling distribution, then the test is a
two – tailed test.
Activity F
2. The probability of committing type I error is
called the level of significance.
3. In a test of hypothesis, a test statistic is computed
from the sample data.
4. The level of significance is set during testing of
hypothesis.
Activity F
5. If we rejected the null hypothesis at 𝛼 = 0.01
then we would also have rejected it at the 𝛼 =
0.05.
6. A type I error is committed when we fail to
reject the null hypothesis when it is actually
true.
Activity F
7. If the computed value falls outside the
rejected region, then the null hypothesis is
rejected.
8. If the symbol less than (<) appears in the
alternative hypothesis then the test is right –
tailed.
Activity F
9. A type II error is committed if the null
hypothesis is rejected when it is false.
10. In a two – tailed test, the rejection region is
split and placed in both tails of the sampling
distribution.
Activity G
Decide whether the null hypothesis is to be
rejected or accepted, given the test value and the
critical value of test statistic. Draw the rejection
region.
Activity G
Probability of Committing Type I or Type II Error
Example. The mean score of all grade 11 students during a
departmental examination in calculus is claimed to be 65
with a standard deviation of 9. The 40 students of section A
have a mean score of 68. The teacher wishes to find out if
the scores of grade 11 students are significantly higher than
the scores of the students in section A. Assuming the scores
are normally distributed,
1. write an appropriate null and alternative
hypotheses; and
2. Compute the probability of committing Type I error.
Solution
Given: 𝜇 = 65 ( population mean)
𝑥ҧ = 68 ( sample mean)
𝜎 = 9 (population standard deviation)
𝑛 = 40
Solution
1. 𝐻𝑜 : The mean score of all grade 11 students
is 65.
𝐻𝑜 : 𝜇 = 65

𝐻𝑎 : The mean score of all grade 11 students

is higher than 65.
𝐻𝑎 : 𝜇 > 65
Solution
2. The probability of committing a Type I error is equivalent to
the probability of getting the alpha error 𝛼. If the scores follow
a normal distribution, then from the Central Limit Theorem:
𝑥ҧ − 𝜇 68 − 65
𝑧= 𝜎 = = 2.11
9
𝑛 40
At z = 2.11, the area under the normal curve to the right of 𝜇 is
0.4826 or 48.26%. Thus, 𝛼=0.0174 or 1.74%.
Hence, the probability of committing a Type I error is 1.74%
Example
The mean age of the registered voters in a certain
municipality is 35. A random sample of 35 registered
voters from the same community has a mean age of
37 and the variance is known to be 36. If the field
reporter wants to prove that the record is not 35,
what is the appropriate null and alternative
hypotheses? What is the probability of getting a Type
I error? Can we conclude that the mean age is not 35
given a confidence level of 95%?
Solution
Given:
𝜇 = 35
𝑥ҧ = 37
2
𝜎 = 36, ; 𝑡ℎ𝑢𝑠, 𝜎 = 6
𝑛 = 35
Solution
𝐻𝑜 : The mean age of the registered voters is 35.
𝐻𝑜 : 𝜇 = 35 or
𝐻𝑜 : There is no significant difference between the mean age of
the population and the sample group registered voters.
𝐻𝑜 : 𝜇 = 𝑥ҧ

𝐻𝑎 : There is a significant difference between the mean age of

the population and the sample group registered voters.
𝐻𝑜 : 𝜇 ≠ 𝑥ҧ
 Solution
The probability of committing a Type I error is equivalent to the
probability of getting the alpha error 𝛼. If the scores follow a normal
distribution, then from the Central Limit Theorem:
𝑥ҧ − 𝜇 37 − 35
𝑧= 𝜎 = = 1.97
6
𝑛 35
At z = 1.97, the value of 𝛼 is equal to 0.0244 or 2.44%.
Hence, the probability of committing a Type I error must be
2.44% × 2 = 4.88%.
Conclusion: We reject the null hypothesis in favor of the alternative
hypothesis.
Take note of the following
❖The conclusion of a hypothesis testing is always
whether the null hypothesis should be rejected
or not. Rejecting the null hypothesis means that
the conclusion is in favor of the alternative
hypothesis.
Take note of the following
❖Experts never accept the null hypothesis. They
always use the term “do not reject” in place of
“accept”
Take note of the following
❖If the p-value is “small”, then the sample data
provides evidence that tends to refute the null
hypothesis; in particular, if the p-value is less than
the significance level 𝛼, then the null hypothesis
can be rejected and the result is statistically
significant at that level. However, if the p-value is
greater than 𝛼, then the null hypothesis is retained
and the result is not statistically significant at that
level.
Solve
The mean water consumption of households living in S&P
Subdivision is 33.5 cubic meters per month. A random
sample of 40 households in the same place has a mean
water consumption of 30 cubic meters per month with a
standard deviation of 8 cubic meters. Formulate an
appropriate null and alternative hypotheses. What is the
probability of committing a Type I error? Can we conclude
that the mean consumption is not 33 cubic meters per
month with a confidence level of 95%? Use a one – tailed
test.
Steps in Hypothesis Testing Using the z - test
❖1. Formulate a null hypothesis 𝐻𝑜 or the
hypothesis to be tested.
❖2. Determine the appropriate test to be used.
We use the z-test if n > 30 and the population
standard deviation is given. On the other hand,
we use t-test if n ≤ 30.
Steps in Hypothesis Testing Using the z - test
❖3. Determine the critical value c using the table
below.
Level of Significance
α = 1% α = 2.5% α = 5% α = 10%
One - tailed c = ±2.33 c = ± 1.96 c = ± 1.645 c = ± 1.28
Two - tailed c = ± 2.575 c = ± 2.33 c = ± 1.96 c = ± 1.645
Steps in Hypothesis Testing Using the z - test
❖4. Compute for the z – score using the Central
Limit Theorem
a. For one sample group
-Standard score of the sample mean using the
population mean:
ҧ
𝑥−𝜇 ҧ
𝑥−𝜇 𝑛
𝑧= 𝜎 or 𝑧 =
𝜎
𝑛
Steps in Hypothesis Testing Using the z - test
b. For two-sample groups with sizes 𝑛1 𝑎𝑛𝑑 𝑛2
*Formulas will be written on the board

5. Decide whether to accept or not to accept 𝐻𝑜.