Hypothesis Testing

John Closter F. Olivo

STAT2206 - Biostatistics

2nd semester, 2019 - 2020

CENTRAL LUZON STATE UNIVERSITY

 Learning Outcomes
DEPARTMENT OF
STATISTICS

After completing this lesson, students will have a more in-depth

understanding about hypothesis testing. Specifically, you would
be able to:

1. Discuss the concept of hypothesis testing;

2. Discuss the importance of hypothesis testing; and

3. Define what a statistical hypothesis is.

Hypothesis Testing !2
 Inferential Statistics
DEPARTMENT OF
STATISTICS

Hypothesis
Estimation
Testing

Point
Estimation

Interval
Estimation

Inferential Statistics !3
 Motivating Example 1
DEPARTMENT OF
STATISTICS

N = 1000 patients

It is claimed that the average weight of all patients in a

particular hospital is 46.8 kgs.

Motivating Example 1 !4
 Motivating Example 1
DEPARTMENT OF
STATISTICS

N = 1000 patients

It is claimed that the average weight of all patients in a

particular hospital is 46.8 kgs.

Motivating Example 1 !5
 Motivating Example 1
DEPARTMENT OF
STATISTICS

N = 1000 patients

It is claimed that the average weight of all patients in a

particular hospital is 46.8 kgs.
µ = 46.8 kgs.
Motivating Example 1 !6
 Motivating Example 1
DEPARTMENT OF
STATISTICS

It is claimed that the average weight of

all patients in a particular hospital is 46.8
N = 1000 patients kgs.

µ = 46.8 kgs.

Are we going to accept or reject the

hypothesis?

Motivating Example 1 !7
 Motivating Example 2
DEPARTMENT OF
STATISTICS

N = 340, 158 households

It is claimed that the proportion of all households with

at least one member aged 90 yrs old is 12%.

Motivating Example 2 !8
 Motivating Example 2
DEPARTMENT OF
STATISTICS

N = 340, 158 households

It is claimed that the proportion of all households with

at least one member aged 90 yrs old is 12%.

Motivating Example 2 !9
 Motivating Example 2
DEPARTMENT OF
STATISTICS

N = 340, 158 households

It is claimed that the proportion of all households with

at least one member aged 90 yrs old is 12%.

p = 0.12 or 12%
Motivating Example 2 !10
 Motivating Example 2
DEPARTMENT OF
STATISTICS

It is claimed that the proportion of all

N = 340, 158 households with at least one member
households
aged 90 yrs old is 12%.

p = 0.12 or 12%

Are we going to accept or reject the

hypothesis?

Motivating Example 2 !11

 Motivating Example 2
DEPARTMENT OF
STATISTICS

What do you observe about

the claims or the
hypotheses?

Motivating Example 2 !12

 Statistical Hypothesis
DEPARTMENT OF
STATISTICS

It is claimed that the average weight of all

N = 1000 patients patients in a particular hospital is 46.8 kgs.

It is claimed that the proportion of all

N = 340, 158
households households with at least one member aged 90
yrs old is 12%.

Motivating Example 2 !13

 Statistical Hypothesis
DEPARTMENT OF
STATISTICS

It is claimed that the average weight of all

N = 1000 patients patients in a particular hospital is 46.8 kgs.

It is an assumption about the

population.

It is claimed that the proportion of households

N = 340, 158
households with at least one member aged 90 yrs old is
12%.

Motivating Example 2 !14

 Hypothesis Testing
DEPARTMENT OF
STATISTICS

This refers to the formal procedures used by

statisticians to accept or reject statistical
hypothesis.

Hypothesis Testing !15

 Hypothesis Testing
DEPARTMENT OF
STATISTICS

N = 1000 patients
It is claimed that the average weight of all
patients in a particular hospital is 46.8 kgs.

Any evidence to support

the claim?

Hypothesis Testing !16

 Hypothesis Testing
DEPARTMENT OF
STATISTICS

N = 340, 158 It is claimed that the proportion of all

households households with at least one member aged
90 yrs old is 12%.

Any evidence to support

the claim?

Hypothesis Testing !17

 Hypothesis Testing
DEPARTMENT OF
STATISTICS

N = 1000 patients n=100 patients

Any evidence to support

the claim?

Hypothesis Testing !18

 Hypothesis Testing
DEPARTMENT OF
STATISTICS

N = 340, 158 n=500

households households

Any evidence to support

the claim?

Hypothesis Testing !19

 Hypothesis Testing
DEPARTMENT OF
STATISTICS

Population
N = 1000
patients

Sampling Probability Sampling Methods

1. Simple Random Sampling

2. Systematic Sampling
3. Stratified Sampling
4. Cluster Sampling
5. Multi-stage Sampling
Sample

n=100
patients

Inference

Hypothesis Testing !20

 Hypothesis Testing
DEPARTMENT OF
STATISTICS

Step 1. Formulate the null and alternative hypotheses.

Step 2. Select the appropriate test statistic and write

the given level of significance.

Step 3. Determine the decision rule. Show the

acceptance and rejection region.

Step 4. Compute the test statistic.

Step 5. Decision.

Step 6. Conclusion.

Hypothesis Testing !21

 Hypothesis Testing
DEPARTMENT OF
STATISTICS

Why do we need to understand the

process of hypothesis testing?

Hypothesis Testing !22

DEPARTMENT OF
STATISTICS Most Commonly Used Statistical
Tests /Analyses
1. t-test

1.1 one-sample t-test

1.2 independent samples t-test

1.3 paired samples t-test

2. Analysis of Variance (ANOVA)

3. Correlation analysis

4. Regression analysis

5. Probit analysis

6. Non-parametric tests

Most Commonly Used Statistical Tests/Analyses !23

 One-Sample t-test
John Closter F. Olivo

STAT2206 - Biostatistics

2nd semester, 2019 - 2020

CENTRAL LUZON STATE UNIVERSITY

 Learning Outcomes
DEPARTMENT OF
STATISTICS

1. Set up null and alternative hypothesis for given problem situations

in both descriptive and symbolic terms.

2. Distinguish between one-tailed and two-tailed tests.

3. Describe the concept of significance level.

4. Determine the critical value of the test statistic for a test.

5. State the decision rule for a test.

6. Compute the sample test statistic and interpret the result.

7. Illustrate Type I and Type II Errors.

One-sample t-test !25

DEPARTMENT OF
STATISTICS
Problem
A researcher claims that the average length at birth for a full-
term baby is 18.35 inches. The length of 30 random sample
newborns was obtained in a a particular hospital. Based on
this sample, the average length was 18.25 with a standard
deviation of 0.98. Test the claim at 5% level of significance.

One-sample t-test !26

 DEPARTMENT OF
STATISTICS Hypothesis Testing Procedures
Step 1. Formulate the null and alternative hypotheses.
1.1 Identify the parameter of interest.

Population mean

Population proportion

Population variance

A researcher claims that the average length at birth for a full-term

baby is 18.35 inches.

1.2 Use the notation for the parameter of interest.

Population mean
µ
Population proportion
p
Population variance s2

Since the parameter of interest is the average we will use the

symbol µ
One-sample t-test !27
 DEPARTMENT OF
STATISTICS Hypothesis Testing Procedures
Step 1. Formulate the null and alternative hypotheses.
1.3 Look for mathematical phrases or keywords.

A researcher claims that the average length at birth for a full-term

baby is 18.35 inches.

µ = 18.35

One-sample t-test !28

 DEPARTMENT OF
STATISTICS Exercises
Write down in notation the claim in the following statements:

1. A researcher claims that the average length at

birth for a full-term baby is at least 18.35 inches.
µ ≥ 18.35
2. A researcher claims that the average length at
birth for a full-term baby is at most 18.35 inches.
µ ≤ 18.35
3. A researcher claims that the average length at birth
for a full-term baby is greater than 18.35 inches.
µ > 18.35
4. A researcher claims that the average length at
birth for a full-term baby is less than 18.35 inches.
µ < 18.35
One-sample t-test !29
 DEPARTMENT OF
STATISTICS Hypothesis Testing Procedures
Step 1. Formulate the null and alternative hypotheses.
1.4 Then, decide if this falls either in null hypothesis or
alternative hypothesis.

Null hypothesis: It always contains the equality sign.

µ = 18.35 µ ≤ 18.35 µ ≥ 18.35

Alternative hypothesis: It always contains the inequality sign.

µ ≠ 18.35 µ > 18.35 µ < 18.35

One-sample t-test !30

 DEPARTMENT OF
STATISTICS Hypothesis Testing Procedures
Null hypothesis: It always contains the equality sign.

µ = 18.35 µ ≤ 18.35 µ ≥ 18.35

Alternative hypothesis: It always contains the inequality sign.

µ ≠ 18.35 µ > 18.35 µ < 18.35

two-tailed test Right-tailed test Left-tailed test

one-tailed test

One-sample t-test !31

 DEPARTMENT OF
STATISTICS Hypothesis Testing Procedures
Step 1. Formulate the null and alternative hypotheses.
1.5 Finally, write down both your null and alternative
hypotheses.

Null hypothesis: µ = 18.35

Alternative hypothesis: µ ≠ 18.35

One-sample t-test !32

 DEPARTMENT OF
STATISTICS Exercise
Formulate the null and alternative hypotheses in the following statements:

1. A researcher claims that the average length at

birth for a full-term baby is at least 18.35 inches.

Null hypothesis: µ ≥ 18.35

Alternative hypothesis: µ < 18.35

2. A researcher claims that the average length at

birth for a full-term baby is at most 18.35 inches.

Null hypothesis: µ ≤ 18.35

Alternative hypothesis: µ > 18.35

One-sample t-test !33

 DEPARTMENT OF
STATISTICS Hypothesis Testing Procedures
Step 2. Select the appropriate test statistic and level of significance.

1.1 Identify the parameters and statistic/s given in the problem.

Parameter Statistic
Size
(# of observations) N n
Mean µ x
Standard deviation σ s

One-sample t-test | !34

DEPARTMENT OF
STATISTICS Hypothesis Testing Procedures
A researcher claims that the average length at birth for a full-
term baby is 18.35 inches. The length of 30 random sample
newborns was obtained in a a particular hospital. Based on
this sample, the average length was 18.25 with a standard
deviation of 0.98. Test the claim at 5% level of significance.

Parameter Statistic

Size Not given 30

Mean 18.35 18.25

Standard deviation Unknown 0.98

One-sample t-test | !35

 DEPARTMENT OF
STATISTICS Hypothesis Testing Procedures

YES Is the YES

POPULATION
Is population normally z-test
STANDARD DEVIATION
distributed?
given (known)?

NO
NO

YES
Is the sample
size large enough? t-test

NO

Non-parametric test
One-sample t-test | !36
 DEPARTMENT OF
STATISTICS Hypothesis Testing Procedures
A researcher claims that the average length at birth for a full-term
baby is 18.35 inches. The length of 30 random sample newborns
was obtained in a a particular hospital. Based on this sample, the
average length was 18.25 with a standard deviation of 0.98. Test the
claim at 5% level of significance. Assume that the population is
normally distributed.

One-sample t-test | !37

 DEPARTMENT OF
STATISTICS Hypothesis Testing Procedures

YES YES
Is the
POPULATION
Is population normally z-test
STANDARD DEVIATION
distributed?
given (known)?

NO
NO

YES
Is the sample
size large enough? t-test

NO

Non-parametric test
One-sample t-test | !38
 DEPARTMENT OF
STATISTICS Hypothesis Testing Procedures
A researcher claims that the average length at birth for a full-term
baby is 18.35 inches. The length of 30 random sample newborns
was obtained in a a particular hospital. Based on this sample, the
average length was 18.25 with a standard deviation of 0.98. Test the
claim at 5% level of significance. Assume that the population is
normally distributed.

Parameter Statistic

Size Not given 30

Mean 18.35 18.25

Standard deviation Unknown 0.98

One-sample t-test | !39

 DEPARTMENT OF
STATISTICS Hypothesis Testing Procedures

YES Is the YES

POPULATION
Is population normally z-test
STANDARD DEVIATION
distributed?
given (known)?

NO
NO

YES
Is the sample
size large enough? t-test

NO

Non-parametric test
One-sample t-test | !40
 DEPARTMENT OF
STATISTICS Hypothesis Testing Procedures
Step 1. Select the appropriate test statistic and level of significance.
Null hypothesis: µ = 18.35
Alternative hypothesis: µ ≠ 18.35
Step 2. Select the appropriate test statistic and level of significance.
Test statistic: t-test
Level of significance: ?

One-sample t-test | !41

 DEPARTMENT OF
STATISTICS Hypothesis Testing Procedures
A researcher claims that the average length at birth for a full-term
baby is 18.35 inches. The length of 30 random sample newborns
was obtained in a a particular hospital. Based on this sample, the
average length was 18.25 with a standard deviation of 0.98. Test the
claim at 5% level of significance. Assume that the population is
normally distributed.

One-sample t-test | !42

 DEPARTMENT OF
STATISTICS Hypothesis Testing Procedures
Step 1. Select the appropriate test statistic and level of significance.
Null hypothesis: µ = 18.35
Alternative hypothesis: µ ≠ 18.35
Step 2. Select the appropriate test statistic and level of significance.
Test statistic: t-test

Level of significance: α = 0.05 or 5%

One-sample t-test | !43

 DEPARTMENT OF
STATISTICS Hypothesis Testing Procedures
Step 3. Determine the decision rule. Show the acceptance and rejection
region.
Alternative Hypothesis

≠ < >
Reject Ho if: Reject Ho if: Reject Ho if:

tc > t α t c < −tα t c > tα

2

One-sample t-test | !44

 DEPARTMENT OF
STATISTICS Hypothesis Testing Procedures
Step 1. Select the appropriate test statistic and level of significance.
Null hypothesis: µ = 18.35
Alternative hypothesis: µ ≠ 18.35
Step 2. Select the appropriate test statistic and level of significance.
Test statistic: t-test
Level of significance: 0.05 or 5%
Step 3. Determine the decision rule. Show the acceptance and rejection

region.

Reject Ho if: tc > t α

,n−1
2

One-sample t-test | !45

 DEPARTMENT OF
STATISTICS Hypothesis Testing Procedures
Step 3. Determine the decision rule. Show the acceptance and rejection

region.

Reject Ho if:
tc > t α
, n−1
2

t c > t 0.05
, 30−1
2

t c > t 0.025, 29
tc > ?

One-sample t-test | !46

 DEPARTMENT OF
STATISTICS Hypothesis Testing Procedures
Step 1:  Look up the
degrees of freedom (df) in
the left hand side of the t-
distribution table.

t c > t 0.025, 29 = ?

One-sample t-test | !47

 DEPARTMENT OF
STATISTICS Hypothesis Testing Procedures
Step 1:  Look up the
degrees of freedom (df) in
the left hand side of the t-
distribution table.

t c > t 0.025, 29 = ?

One-sample t-test | !48

 DEPARTMENT OF
STATISTICS Hypothesis Testing Procedures

Step 2: Locate the column

under your alpha level(the
alpha level is usually given
to you in the question).

t c > t 0.025, 29 = ?

One-sample t-test | !49

 DEPARTMENT OF
STATISTICS Hypothesis Testing Procedures

Step 2: Locate the column

under your alpha level(the
alpha level is usually given
to you in the question).

t c > t 0.025, 29 = 2.04523

One-sample t-test | !50

 DEPARTMENT OF
STATISTICS Hypothesis Testing Procedures
Step 3. Determine the decision rule. Show the acceptance and rejection

region.

t c > 2.04523

Acceptance Region

Rejection
Region Rejection
Region

-2.0453 2.0453
One-sample t-test | !51
 DEPARTMENT OF
STATISTICS Hypothesis Testing Procedures
Step 1. Select the appropriate test statistic and level of significance.
Null hypothesis: µ = 18.35
Alternative hypothesis: µ ≠ 18.35
Step 2. Select the appropriate test statistic and level of significance.

Test statistic: t-test

α = 0.05 or 5%
α

Level of significance:

Step 3. Determine the decision rule. Show the acceptance and rejection

region.

t c > 2.04523

One-sample t-test | !52

 DEPARTMENT OF
STATISTICS Hypothesis Testing Procedures
Step 4. Compute the test statistic.

t c > 2.04523

x −µ
tc =
s
n

One-sample t-test | !53

DEPARTMENT OF
STATISTICS Hypothesis Testing Procedures
x −µ
tc =
s
n
Parameter Statistic

Size Not given 30

Mean 18.35 18.25

Standard deviation Unknown 0.98

x − µ 18.25−18.35
tc = = = −0.56
s 0.98
n 30
One-sample t-test | !54
 DEPARTMENT OF
STATISTICS Hypothesis Testing Procedures
Step 1. Select the appropriate test statistic and level of significance.
Null hypothesis: µ = 18.35
Alternative hypothesis: µ ≠ 18.35
Step 2. Select the appropriate test statistic and level of significance.
Test statistic: t-test
Level of significance: α = 0.05 or 5%
Step 3. Determine the decision rule. Show the acceptance

and rejection region. α

Reject Ho if: t c > 2.04523

Step 4. Compute the test statistic.

x − µ 18.25−18.35
tc = = = −0.56
s 0.98
n 30

One-sample t-test | !55

 DEPARTMENT OF
STATISTICS Hypothesis Testing Procedures
Step 5. Decision.

Reject Ho if: t c > 2.04523

x − µ 18.25−18.35
tc = = = −0.56
s 0.98
n 30
−0.56 > 2.04523
0.56 < 2.04523
Since 0.56 is less than 2.04523, we fail to reject the null hypothesis.

Remarks:
1) Reject Ho if the test statistic falls in the critical region
2) Fail to Reject Ho if the test statistic does NOT fall in the critical region.

One-sample t-test | !56

 DEPARTMENT OF
STATISTICS Hypothesis Testing Procedures
Step 1. Select the appropriate test statistic and level of significance.
Null hypothesis: µ = 18.35
Alternative hypothesis: µ ≠ 18.35
Step 2. Select the appropriate test statistic and level of significance.
Test statistic: t-test
Level of significance: α = 0.05 or 5%
Step 3. Determine the decision rule. Show the acceptance

and rejection region. α

Reject Ho if: t c > 2.04523

Step 4. Compute the test statistic.
x − µ 18.25−18.35
tc = = = −0.56
s 0.98
n 30
Step 5. Decision.
Fail to reject the null hypothesis.

Step 6. Conclusion.

One-sample t-test | !57

 DEPARTMENT OF
STATISTICS Hypothesis Testing Procedures
Step 6. Conclusion.

If the CLAIM is…

Null Alternative
Hypothesis Hypothesis
(Ho) (Ha)

Reject Ho. Fail to reject Ho. Reject Ho. Fail to reject Ho.

“There is sufficient “There is not “There is sufficient “There is not

evidence to reject the sufficient evidence to evidence to support sufficient to support
claim that…” reject the claim the claim that …” the claim that …”

One-sample t-test | !58

 DEPARTMENT OF
STATISTICS Problem
A researcher claims that the average length at birth for a full-term
baby is 18.35 inches. The length of 30 random sample newborns
was obtained in a a particular hospital. Based on this sample, the
average length was 18.25 with a standard deviation of 0.98. Test the
claim at 5% level of significance. Assume that the population is
normally distributed.

Null hypothesis: µ = 18.35

Decision: Fail to reject the null hypothesis.

One-sample t-test | !59

 DEPARTMENT OF
STATISTICS Hypothesis Testing Procedures
Step 6. Conclusion.

If the CLAIM is…

Null Alternative
Hypothesis Hypothesis
(Ho) (Ha)

Reject Ho. Fail to reject Ho. Reject Ho. Fail to reject Ho.

“There is sufficient “There is not “There is sufficient “There is not

evidence to reject the sufficient evidence to evidence to support sufficient to support
claim that…” reject the claim the claim that …” the claim that …”

One-sample t-test | !60

 DEPARTMENT OF
STATISTICS Hypothesis Testing Procedures
Step 6. Conclusion.

If the CLAIM is…

Null Alternative
Hypothesis Hypothesis
(Ho) (Ha)

Reject Ho. Fail to reject Ho. Reject Ho. Fail to reject Ho.

“There is sufficient “There is not “There is sufficient “There is not

evidence to reject the sufficient evidence to evidence to support sufficient to support
claim that…” reject the claim the claim that …” the claim that …”

One-sample t-test | !61

 DEPARTMENT OF
STATISTICS Hypothesis Testing Procedures
Step 6. Conclusion.

If the CLAIM is…

Null Alternative
Hypothesis Hypothesis
(Ho) (Ha)

Reject Ho. Fail to reject Ho. Reject Ho. Fail to reject Ho.

“There is sufficient “There is not “There is sufficient “There is not

evidence to reject the sufficient evidence to evidence to support sufficient to support
claim that…” reject the claim the claim that …” the claim that …”

One-sample t-test | !62

 DEPARTMENT OF
STATISTICS Hypothesis Testing Procedures
Step 6. Conclusion.

If the CLAIM is…

Null Alternative
Hypothesis Hypothesis
(Ho) (Ha)

Reject Ho. Fail to reject Ho. Reject Ho. Fail to reject Ho.

“There is sufficient “There is not “There is sufficient “There is not

evidence to reject the sufficient evidence to evidence to support sufficient to support
claim that…” reject the claim the claim that …” the claim that …”

At 5% level of significance, there is not sufficient evidence to reject the

claim that the average length at birth for a full-term baby is 18.35 inches.
One-sample t-test | !63
 DEPARTMENT OF
STATISTICS Hypothesis Testing Procedures
Step 1. Select the appropriate test statistic and level of significance.
Null hypothesis: µ = 18.35
Alternative hypothesis: µ ≠ 18.35
Step 2. Select the appropriate test statistic and level of significance.
Test statistic: t-statistic
Level of significance: α = 0.05 or 5%
Step 3. Determine the decision rule. Show the acceptance

and rejection region. α

Reject Ho if: t c > 2.04523

Step 4. Compute the test statistic.
x − µ 18.25−18.35
tc = = = −0.56
s 0.98
n 30
Step 5. Decision.
Fail to reject the null hypothesis.

Step 6. Conclusion.
At 5% level of significance, there is not sufficient evidence to reject the claim that the average
length at birth for a full-term baby is 18.35 inches.
One-sample t-test | !64
DEPARTMENT OF
STATISTICS Problem
A researcher claims that the mean height of women ages 20
and older is greater than 62 inches. A random sample of 27
women ages 20 and older was selected. Based on this sample,
the average height was 62.10 inches with a standard deviation
of 0.20. Test the claim at 5% level of significance. Assume that
the sample comes from a normally distributed population.

One-sample t-test | !65

 DEPARTMENT OF
STATISTICS Hypothesis Testing Procedures
Step 1. Formulate the null and alternative hypotheses.
1.1 Identify the parameter of interest.

Population mean

Population proportion

Population variance

A researcher claims that the mean height of women ages 20 and

older is greater than 62 inches.

1.2 Use the notation for the parameter of interest.

Population mean
µ
Population proportion
p
Population variance s2

Since the parameter of interest is the mean or average we will

use the symbol µ
One-sample t-test | !66
 DEPARTMENT OF
STATISTICS Hypothesis Testing Procedures
Step 1. Formulate the null and alternative hypotheses.
1.3 Look for mathematical phrases or keywords.

A researcher claims that the mean height of women ages 20 and

older is greater than 62 inches.

µ > 62

One-sample t-test | !67

 DEPARTMENT OF
STATISTICS Hypothesis Testing Procedures
Step 1. Formulate the null and alternative hypotheses.
1.4 Then, decide if this falls either null hypothesis or
alternative hypothesis.

Alternative hypothesis: µ > 62

One-sample t-test | !68

 DEPARTMENT OF
STATISTICS Hypothesis Testing Procedures
Step 1. Formulate the null and alternative hypotheses.
1.5 Finally, write down both your null and alternative
hypotheses.

Null hypothesis: µ ≤ 62
Alternative hypothesis: µ > 62

One-sample t-test | !69

 DEPARTMENT OF
STATISTICS Hypothesis Testing Procedures
Step 2. Select the appropriate test statistic and level of significance.

1.1 Identify the parameters and statistic/s given in the problem.

Parameter Statistic
Size
(# of observations) N n
Mean µ x
Standard deviation σ s

One-sample t-test | !70

DEPARTMENT OF
STATISTICS Hypothesis Testing Procedures
A researcher claims that the mean height of women ages 20
and older is greater than 62 inches. A random sample of 27
women ages 20 and older was selected. Based on this sample,
the average height was 62.10 inches with a standard deviation
of 0.20. Test the claim at 5% level of significance. Assume that
the sample comes from a normally distributed population.

Parameter Statistic

Size Not given 27

Mean 62 62.10

Standard deviation Unknown 0.20

One-sample t-test | !71

 DEPARTMENT OF
STATISTICS Hypothesis Testing Procedures

YES Is the YES

POPULATION
Is population normally z-test
STANDARD DEVIATION
distributed?
given (known)?

NO
NO

YES
Is the sample
size large enough? t-test

NO

Non-parametric test

One-sample t-test | !72

DEPARTMENT OF
STATISTICS Hypothesis Testing Procedures
A researcher claims that the mean height of women ages 20
and older is greater than 62 inches. A random sample of 27
women ages 20 and older was selected. Based on this sample,
the average height was 62.10 inches with a standard deviation
of 0.20. Test the claim at 5% level of significance. Assume that
the sample comes from a normally distributed population.

Parameter Statistic

Size Not given 27

Mean 62 62.10

Standard deviation Unknown 0.20

One-sample t-test | !73

 DEPARTMENT OF
STATISTICS Hypothesis Testing Procedures

YES Is the YES

POPULATION
Is population normally z-test
STANDARD DEVIATION
distributed?
given (known)?

NO
NO

YES
Is the sample
size large enough? t-test

NO

Non-parametric test
One-sample t-test | !74
DEPARTMENT OF
STATISTICS Hypothesis Testing Procedures
A researcher claims that the mean height of women ages 20
and older is greater than 62 inches. A random sample of 27
women ages 20 and older was selected. Based on this sample,
the average height was 62.10 inches with a standard deviation
of 0.20. Test the claim at 5% level of significance. Assume that
the sample comes from a normally distributed population.

Parameter Statistic

Size Not given 27

Mean 62 62.10

Standard deviation Unknown 0.20

One-sample t-test | !75

 DEPARTMENT OF
STATISTICS Hypothesis Testing Procedures

YES Is the YES

POPULATION
Is population normally z-test
STANDARD DEVIATION
distributed?
given (known)?

NO
NO

YES
Is the sample
size large enough? t-test

NO

Non-parametric test

One-sample t-test | !76

 DEPARTMENT OF
STATISTICS Hypothesis Testing Procedures
Step 1. Select the appropriate test statistic and level of significance.
Null hypothesis: µ ≤ 62
Alternative hypothesis: µ > 62
Step 2. Select the appropriate test statistic and level of significance.
Test statistic: t-test
Level of significance: 0.05 or 5%

One-sample t-test | !77

 DEPARTMENT OF
STATISTICS Hypothesis Testing Procedures
Step 3. Determine the decision rule. Show the acceptance and rejection
region.
Alternative Hypothesis

≠ < >
Reject Ho if: Reject Ho if: Reject Ho if:

tc > t α t c < −tα t c > tα

2

One-sample t-test | !78

 DEPARTMENT OF
STATISTICS Hypothesis Testing Procedures
Step 1. Select the appropriate test statistic and level of significance.
Null hypothesis: µ ≤ 62
Alternative hypothesis: µ > 62

Step 2. Select the appropriate test statistic and level of significance.

Test statistic: t-test
Level of significance: 0.05 or 5%
Step 3. Determine the decision rule. Show the acceptance and rejection

region.

Reject Ho if:

t c > tα

One-sample t-test | !79

 DEPARTMENT OF
STATISTICS Hypothesis Testing Procedures
Step 3. Determine the decision rule. Show the acceptance and rejection

region.

Reject Ho if:

t c > tα ,n−1
t c > t 0.05, 27−1
t c > t 0.05, 26
tc > ?

One-sample t-test | !80

 DEPARTMENT OF
STATISTICS Hypothesis Testing Procedures
Step 1:  Look up the
degrees of freedom (df) in
the left hand side of the t-
distribution table.

t c > t 0.05, 26 = ?

One-sample t-test | !81

 DEPARTMENT OF
STATISTICS Hypothesis Testing Procedures
Step 1:  Look up the
degrees of freedom (df) in
the left hand side of the t-
distribution table.

t c > t 0.05, 26 = ?

One-sample t-test | !82

 DEPARTMENT OF
STATISTICS Hypothesis Testing Procedures

Step 2: Locate the column

under your alpha level(the
alpha level is usually given
to you in the question).

t c > t 0.05, 26 = ?

One-sample t-test | !83

 DEPARTMENT OF
STATISTICS Hypothesis Testing Procedures

Step 2: Locate the column

under your alpha level(the
alpha level is usually given
to you in the question).

t c > t 0.05, 26 = 1.7056

One-sample t-test | !84

 DEPARTMENT OF
STATISTICS Hypothesis Testing Procedures
Step 3. Determine the decision rule. Show the acceptance and rejection

region.

t c > 1.7056

Acceptance Region

Rejection
Region

1.7056
One-sample t-test | !85
 DEPARTMENT OF
STATISTICS Hypothesis Testing Procedures
Step 1. Select the appropriate test statistic and level of significance.
Null hypothesis: µ ≤ 62
Alternative hypothesis: µ > 62

Step 2. Select the appropriate test statistic and level of significance.

Test statistic: t-test

α = 0.05 or 5%
α

Level of significance:

Step 3. Determine the decision rule. Show the acceptance and rejection

region.

t c > 1.7056

One-sample t-test | !86

 DEPARTMENT OF
STATISTICS Hypothesis Testing Procedures
Step 4. Compute the test statistic.

t c > 1.7056

x −µ
tc =
s
n

One-sample t-test | !87

DEPARTMENT OF
STATISTICS Hypothesis Testing Procedures
x −µ
tc =
s
n

Parameter Statistic

Size Not given 27

Mean 62 62.10

Standard deviation Unknown 0.20

x − µ 62.10− 62
tc = = = 2.60
s 0.20
n 27
One-sample t-test | !88
 DEPARTMENT OF
STATISTICS Hypothesis Testing Procedures
Step 1. Select the appropriate test statistic and level of significance.
Null hypothesis: µ ≤ 62
Alternative hypothesis: µ > 62
Step 2. Select the appropriate test statistic and level of significance.
Test statistic: t-test
Level of significance: α = 0.05 or 5%
Step 3. Determine the decision rule. Show the acceptance

and rejection region. α

Reject Ho if: t c > 1.7056

Step 4. Compute the test statistic.

x − µ 62.10− 62
tc = = = 2.60
s 0.20
n 27
One-sample t-test | !89
 DEPARTMENT OF
STATISTICS Hypothesis Testing Procedures
Step 5. Decision.
Reject Ho if: t c > 1.7056

x − µ 62.10− 62
tc = = = 2.60
s 0.20
n 27

2.60 > 1.7056

Since 2.60 is greater than 1.7056, we reject the null hypothesis.

One-sample t-test | !90

 DEPARTMENT OF
STATISTICS Hypothesis Testing Procedures
Step 6. Conclusion.

If the CLAIM is…

Null Alternative
Hypothesis Hypothesis
(Ho) (Ha)

Reject Ho. Fail to reject Ho. Reject Ho. Fail to reject Ho.

“There is sufficient “There is not “There is sufficient “There is not

evidence to reject the sufficient evidence to evidence to support sufficient to support
claim that…” reject the claim the claim that …” the claim that …”

One-sample t-test | !91

DEPARTMENT OF
STATISTICS Problem
A researcher claims that the mean height of women ages 20
and older is greater than 62 inches. A random sample of 27
women ages 20 and older was selected. Based on this sample,
the average height was 62.10 inches with a standard deviation
of 0.20. Test the claim at 5% level of significance. Assume that
the sample comes from a normally distributed population.

Alternative hypothesis: µ > 62

Decision: Reject the null hypothesis.

One-sample t-test | !92

 DEPARTMENT OF
STATISTICS Hypothesis Testing Procedures
Step 6. Conclusion.

If the CLAIM is…

Null Alternative
Hypothesis Hypothesis
(Ho) (Ha)

Reject Ho. Fail to reject Ho. Reject Ho. Fail to reject Ho.

“There is sufficient “There is not “There is sufficient “There is not

evidence to reject the sufficient evidence to evidence to support sufficient to support
claim that…” reject the claim the claim that …” the claim that …”

One-sample t-test | !93

 DEPARTMENT OF
STATISTICS Hypothesis Testing Procedures
Step 6. Conclusion.

If the CLAIM is…

Null Alternative
Hypothesis Hypothesis
(Ho) (Ha)

Reject Ho. Fail to reject Ho. Reject Ho. Fail to reject Ho.

“There is sufficient “There is not “There is sufficient “There is not

evidence to reject the sufficient evidence to evidence to support sufficient to support
claim that…” reject the claim the claim that …” the claim that …”

One-sample t-test | !94

 DEPARTMENT OF
STATISTICS Hypothesis Testing Procedures
Step 6. Conclusion.

If the CLAIM is…

Null Alternative
Hypothesis Hypothesis
(Ho) (Ha)

Reject Ho. Fail to reject Ho. Reject Ho. Fail to reject Ho.

“There is sufficient “There is not “There is sufficient “There is not

evidence to reject the sufficient evidence to evidence to support sufficient to support
claim that…” reject the claim the claim that …” the claim that …”

One-sample t-test | !95

 DEPARTMENT OF
STATISTICS Hypothesis Testing Procedures
Step 6. Conclusion.

If the CLAIM is…

Null Alternative
Hypothesis Hypothesis
(Ho) (Ha)

Reject Ho. Fail to reject Ho. Reject Ho. Fail to reject Ho.

“There is sufficient “There is not “There is sufficient “There is not

evidence to reject the sufficient evidence to evidence to support sufficient to support
claim that…” reject the claim the claim that …” the claim that …”

At 5% level of significance, there is sufficient evidence to support the

claim that the mean height of women ages 20 and older is greater than
62 inches.
One-sample t-test | !96
 DEPARTMENT OF
STATISTICS Hypothesis Testing Procedures
Step 1. Select the appropriate test statistic and level of significance.
Null hypothesis: µ ≤ 0.87
Alternative hypothesis: µ > 0.87
Step 2. Select the appropriate test statistic and level of significance.
Test statistic: t-test
Level of significance: α = 0.05 or 5%
Step 3. Determine the decision rule. Show the acceptance

and rejection region. α

Reject Ho if: t c > 1.7056

Step 4. Compute the test statistic.
x − µ 62.10− 62
tc = = = 2.60
s 0.20
n 27
Step 5. Decision.
Reject the null hypothesis.

Step 6. Conclusion.
At 5% level of significance, there is sufficient evidence to support the claim that the that the
mean height of women ages 20 and older is greater than 62 inches.
One-sample t-test | !97
DEPARTMENT OF
STATISTICS t-table

One-sample t-test | !98

DEPARTMENT OF
STATISTICS Types of Error

Rejecting the null hypothesis when in fact

it is true.
Type I Error
The probability of committing a Type I
error is alpha.

Accepting the null when in fact it is false.

Type II Error
The probability of committing a Type II
error is beta.

One-sample t-test | !99