Hypothesis Testing

Name it!
Small letter sigma

myu Not equal
estimate
Normal distribution graph

sample
Hypothesis Testing for a Population Mean
• How to use the sample mean to estimate the population
mean?
Average age
Average age of the
of estimate
entire CNHS
JS in CNHS
• We use three techniques to estimate the population mean
Point Confidence Hypothesis

Estimate Interval Testing
Population √ √
Mean
• The hypotheses testing is like a debate.
The common characteristics are :

μ must
μ must 1) In a Hypotheses Testing, we always
be less
be 125 than 125 have two sides of opinions about the
value of the population parameter.
2) The support of the Hypotheses Testing

is the observed sample information
3) According to some standard, we will

make a decision at the end of the
Hypotheses Testing about which side
wins!
Definition of terms:
• Hypothesis – is a tentative presupposition or

inference made in order to predict the
occurrence of a phenomenon.
• Hypothesis testing – is a procedure in making

decisions based on a sample evidence
used to determine whether the
hypothesis is accepted or rejected.
The whole reasoning process of a Hypotheses Testing
includes the following steps:
Step 1: Set up appropriate hypotheses
Step 2: Check the assumptions
Step 3: Compute the test statistic
Step 4: Make a decision

• Step 1 Set up appropriate hypotheses
– Two sides of opinions:
the null hypothesis and the alternative hypothesis.
– The null hypothesis, which we denote H0 , specifies that there is no

difference between a parameter and a specific value.
For example : H0 : μ=125
– The alternative hypothesis, which we denote HA , states the existence

of a difference between a parameter and a specific value. For example :
H1 : μ <125 or H1 : μ >125 or H1 : μ ≠125
Step 1 Set up appropriate hypotheses
Now we can transform the comic into :
μ must
μ must not be H1 : μ ≠ 125
H0: μ =125
be 125 equal to
125
3 forms of hypotheses for population mean
Null Hypothesis H0 vs Alternative Hypothesis Ha
H0: μ = μ0 Ha : μ>μ0 (one-tailed)

Ha : μ<μ0 (one-tailed)
Ha : μ≠μ0 (two-tailed)
Question:
How to choose an appropriate form of hypotheses for a
particular case?
In this course,
1) H0 is always stated using the equals sign.

2) H1 always uses a “not equal sign”: <, >, ≠,
< : Is less /short than, Is decreased /reduced from, …

> : Is greater/higher/longer/than, Is increased,…
≠ : Is not equal to, Is not the same as, Has changed from,…
Formulating Hypothesis
1. The average TV viewing time of all five-year old children is 4

hours daily.
2. The mean performance of all grade six leavers of a school in the

NAT is 35.
3. The inventor of a new kind of light bulb claims that all such
bulbs last less than 3000 hrs.
• Example 1
A researcher wishes to test the claim that the average cost of tuition
and fees at a four-year public college is greater than $5700. She
selects a random sample of 36 four-year public colleges and finds
the mean to be $5950. The population standard deviation is $659.
Question Please set up appropriate hypotheses.
Solution H0 : μ = 5950 , H1 : μ > 5950


• Example 1
A researcher wishes to test the claim that the average cost of

tuition and fees at a four-year public college is greater than $5700.
She selects a random sample of 36 four-year public colleges and
finds the mean to be P5950. The population standard deviation is
P659.
• Questions
•Are the samples selected randomly? yes

• Is the sample size large enough for the CLT to apply? yes
• What type of test statistic we are going to use? z test

• Example 1

P659. Use the 95% confidence level.
Compute the test statistic, denoted by z
Question Please calculate the test statistics. Z = 2.28


• Example 1

P659.
Choose a significance level for α :
α = .05
The decision to reject or fail to reject the null hypothesis is

called significance and it should be based on a set of
criterion of judgment called the level of significance.
Step4: Make a decision
If the p-value is low enough, we prefer to reject H0

Or
If the alpha level α is given,
when p-value< α , we reject the null hypothesis
when p-value> α , we do not reject the null hypothesis
P-value is the probability that the observed sample statistical value (

or an even more extreme value) occur if the null model H0 is true.
Alternative Hypothesis HA P-value formula

Ha : μ≠μ0 (two-sided) p-value=2P(Z>|z|)
Ha : μ>μ0 (one-sided) p-value=P(Z>z)
HA : μ<μ0 (one-sided) p-value=P(Z<z)

• Example 1

P659.
Question
Is there evidence to support the claim at α = .05 ?
Since the z = 2.28, its equivalent area is 0.4887

Now, p value = p (z > computed z)
= 0.5000 – 0.4887
= 0.0113
Table of Z
• Interpret the result.
We know that 0.0113 < 0.05.
Thus, based on the evidence at hand, we reject the null hypothesis.
Therefore, there is a significant difference

between the sample mean and population mean.

Hypothesis Testing

Uploaded by

Copyright:

Available Formats

Hypothesis Testing

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Hypothesis Testing

Uploaded by

Copyright:

Available Formats

Name it!

Small letter sigma

Normal distribution graph

• We use three techniques to estimate the population mean

Point Confidence Hypothesis

The common characteristics are :

2) The support of the Hypotheses Testing

3) According to some standard, we will

• Hypothesis – is a tentative presupposition or

• Hypothesis testing – is a procedure in making

Step 1: Set up appropriate hypotheses

Step 2: Check the assumptions

Step 3: Compute the test statistic

Step 4: Make a decision

– Two sides of opinions:

the null hypothesis and the alternative hypothesis.

– The null hypothesis, which we denote H0 , specifies that there is no

– The alternative hypothesis, which we denote HA , states the existence

Now we can transform the comic into :

3 forms of hypotheses for population mean

Null Hypothesis H0 vs Alternative Hypothesis Ha

H0: μ = μ0 Ha : μ>μ0 (one-tailed)

1) H0 is always stated using the equals sign.

< : Is less /short than, Is decreased /reduced from, …

1. The average TV viewing time of all five-year old children is 4

2. The mean performance of all grade six leavers of a school in the

Question Please set up appropriate hypotheses.

Solution H0 : μ = 5950 , H1 : μ > 5950

Step 1: Set up appropriate hypotheses

Step 2: Check the assumptions

Step 3: Compute the test statistic

Step 4: Make a decision

A researcher wishes to test the claim that the average cost of

•Are the samples selected randomly? yes

Step 1: Set up appropriate hypotheses

Step 2: Check the assumptions

Step 3: Compute the test statistic

Step 4: Make a decision

A researcher wishes to test the claim that the average cost of

Compute the test statistic, denoted by z

Question Please calculate the test statistics. Z = 2.28

Step 1: Set up appropriate hypotheses

Step 2: Check the assumptions

Step 3: Compute the test statistic

Step 4: Make a decision

A researcher wishes to test the claim that the average cost of

Choose a significance level for α :

The decision to reject or fail to reject the null hypothesis is

If the p-value is low enough, we prefer to reject H0

P-value is the probability that the observed sample statistical value (

Alternative Hypothesis HA P-value formula

Ha : μ>μ0 (one-sided) p-value=P(Z>z)

HA : μ<μ0 (one-sided) p-value=P(Z<z)

A researcher wishes to test the claim that the average cost of

Since the z = 2.28, its equivalent area is 0.4887

We know that 0.0113 < 0.05.

Thus, based on the evidence at hand, we reject the null hypothesis.

Therefore, there is a significant difference