Kwame Nkrumah University

of Science and Technology

NUT 561
Statistical Methods and Biostatistics

Dr. Emmanuel de-Graft Johnson Owusu-Ansah

[email protected]
0244378150

© 2014 Cengage Learning. All Rights Reserved. May not be scanned, copied
Slide 1 1
or duplicated, or posted to a publicly accessible website, in whole or in part.
Inference About Means, Proportions and Variances
with Two Populations

 Inferences About the Difference Between

Two Population Variances

 Inferences About the Difference Between

Two Population Means: σ known

 Inferences About the Difference Between

Two Population Means: σ unknown

 Inferences About the Difference Between

Means: Paired Samples

 Inferences About the Difference Between

Two Population Proportions

© 2014 Cengage Learning. All Rights Reserved. May not be scanned, copied
Slide 2
or duplicated, or posted to a publicly accessible website, in whole or in part.
 Two-Tailed Hypothesis Testing About the
Variances of Two Populations

H 0 :  12   22
• Hypotheses
Ha : 12   22

• Test Statistic s12

F 2
s2

• The test statistic follows an F distribution with

degrees of freedom equals n1-1, n2-1
• Rejection Rule
Reject H0 if p-value < a
© 2014 Cengage Learning. All Rights Reserved. May not be scanned, copied
Slide 3
or duplicated, or posted to a publicly accessible website, in whole or in part.
 Hypothesis Testing About the
Variances of Two Populations
 One-Tailed Test
• Hypotheses
H 0 :  12   22
H a :  12   22

• Test Statistic
s12
F 2
s2

• Rejection Rule
Reject H0 if p-value < a

© 2014 Cengage Learning. All Rights Reserved. May not be scanned, copied
Slide 4
or duplicated, or posted to a publicly accessible website, in whole or in part.
 Two-Tailed Hypothesis Testing About the
Variances of Two Populations (Levene’s Test)

• Hypotheses H 0 :  12   22
Ha : 12   22

Given a variable Y with sample size N divided into k

subgroups, where Ni is the sample size of the ith subgroup,
the Levene test statistic is defined as:

© 2014 Cengage Learning. All Rights Reserved. May not be scanned, copied
Slide 5
or duplicated, or posted to a publicly accessible website, in whole or in part.
 Two-Tailed Hypothesis Testing About the
Variances of Two Populations (Levene’s Test)

The test statistic W has an F distribution with k-1 and N-k degrees of freedom
© 2014 Cengage Learning. All Rights Reserved. May not be scanned, copied
Slide 6
or duplicated, or posted to a publicly accessible website, in whole or in part.
 Hypothesis Testing About the
Variances of Two Populations
Example: Weight of malnourished children
A nutritionist claims that the variance of the weight of
malnourished female children is different from males. A
recent study obtained the weight of 20 random sample
malnourished children and the data is shown below.
We will conduct a hypothesis test with a = 0.05 to see if the
variances are equal for male and female

Female. 14.4 16.8 16.1 17.3 16.1 11.3 16.6 14.8 14.9 16.3
Male. 18.1 16.3 10.4 15.5 14.3 13.9 13.5 10.8 15.1 12.8

© 2014 Cengage Learning. All Rights Reserved. May not be scanned, copied
Slide 7
or duplicated, or posted to a publicly accessible website, in whole or in part.
 Hypothesis Testing About the
Variances of Two Populations

H0: σ2F = σ2M

1. Develop the hypotheses. Ha: σ2F ≠ σ2M

2. Specify the level of significance. a = 0.05

3. Compute the value of the Levene’s test statistic

Conduct the analysis in SPSS as shown on the next slide

© 2014 Cengage Learning. All Rights Reserved. May not be scanned, copied
Slide 8
or duplicated, or posted to a publicly accessible website, in whole or in part.
 Hypothesis Testing About the
Variances of Two Populations

SPSS Output

© 2014 Cengage Learning. All Rights Reserved. May not be scanned, copied
Slide 9
or duplicated, or posted to a publicly accessible website, in whole or in part.
 Hypothesis Testing About the
Variances of Two Populations

4. Compute the p –value.

For W = 0.781, the corresponding p-value is 0.389

5. Determine whether to reject H0.

Since the p–value = 0.389 is greater than a = 0.05,

we fail to reject H0.
There is no sufficient evidence to support the claim that the
variance of the weight of malnourished female children is
different from males
© 2014 Cengage Learning. All Rights Reserved. May not be scanned, copied
Slide 10
or duplicated, or posted to a publicly accessible website, in whole or in part.
 Hypothesis Tests About m 1 - m 2:
σ known
 Hypotheses

H 0 : m1 - m2  D0 H 0 : m1 - m2  D0 H 0 : m1 - m2  D0
H a : m1 - m2  D0 H a : m1 - m2  D0 H a : m1 - m2  D0
Left-tailed Right-tailed Two-tailed
 Test Statistic
( x1 - x2 ) - D0
z
 12  22

n1 n2

This test statistic has a standard normal probability Distribution z

© 2014 Cengage Learning. All Rights Reserved. May not be scanned, copied
Slide 11
or duplicated, or posted to a publicly accessible website, in whole or in part.
 Hypothesis Tests About m 1 - m 2:
σ unknown (Assume Equal variances)
 Hypotheses

H 0 : m1 - m2  D0 H 0 : m1 - m2  D0 H 0 : m1 - m2  D0
H a : m1 - m2  D0 H a : m1 - m2  D0 H a : m1 - m2  D0
Left-tailed Right-tailed Two-tailed

 Test Statistic

( x1 - x2 ) - ( m1 - m 2 )
t -  -
2 2
( n 1)s ( n 1)s
s2  1 1 2 2
s2 (1 n1  1 n2 ) n1  n2 - 2

This test statistic has a t distribution with

n1+ n2 - 2 degrees of freedom. Slide 12
 Hypothesis Tests About m 1 - m 2:
σ unknown (Assume Unequal variances)
 Hypotheses

H 0 : m1 - m2  D0 H 0 : m1 - m2  D0 H 0 : m1 - m2  D0
H a : m1 - m2  D0 H a : m1 - m2  D0 H a : m1 - m2  D0
Left-tailed Right-tailed Two-tailed

 Test Statistic Where the degrees of freedom for ta/2 is:

( x1 - x2 ) - D0
2
s s 
2 2
t   
1 2

s12 s22
df   n1 n2 
 2 2
n1 n2 1  s1 2
1  s2 
2

    
n1 - 1  n1  n2 - 1  n2 
Slide 13
 Hypothesis Tests About m 1 - m 2:
σ unknown (Assume Equal variances)
Example: Average weight of malnourished children
A nutritionist claims that the average weight of
malnourished female children is different from males. A
recent study obtained the weight of 20 random sample
malnourished children and the data is shown below.
We will conduct a hypothesis test with a = 0.05 to test the
claim

Female. 14.4 16.8 16.1 17.3 16.1 11.3 16.6 14.8 14.9 16.3
Male. 18.1 16.3 10.4 15.5 14.3 13.9 13.5 10.8 15.1 12.8

Slide 14
 Hypothesis Tests About m 1 - m 2:
σ unknown (Assume Equal variances)

H0: mM = mF
1. Develop the hypotheses. Ha: mM ≠ mF

2. Specify the level of significance. a = 0.05

3. Compute the value of the test statistic.

Conduct the analysis in SPSS as shown on the next slide

Slide 15
 Hypothesis Tests About m 1 - m 2:
σ unknown (Assume Equal variances)

SPSS Output

Slide 16
 Hypothesis Tests About m 1 - m 2:
σ unknown (Assume Equal variances)

4. Compute the p –value.

For z = 1.496, the corresponding p-value is 0.152

5. Determine whether to reject H0.

Since the p–value = 0.152 is greater than a = 0.05,

we fail to reject H0.

There is no sufficient evidence to support the claim

that the average weight of malnourished female
children is different from males

Slide 17
 Hypothesis Tests About m 1 - m 2:
σ unknown (Assume Unequal Variances)
Example: Carbohydrate
A dietitian claims that on average the carbohydrate intake
by obese pregnant women is different from that of non-
obese pregnant women. A recent experiment involving a
random sample of 10 obese and 10 non-obese pregnant
women resulted in the data below. Test the claim of the
dietitian at 95% confidence level

Obese 32 38 37 36 65 34 39 53 37 42
Non-obese 30 36 35 36 31 34 37 33 32 30

Slide 18
 Hypothesis Tests About m 1 - m 2:
σ unknown (Assume Equal variances)

H0: mY = mN
1. Develop the hypotheses. Ha: mY ≠ mN

2. Specify the level of significance. a = 0.05

3. Compute the value of the test statistic.

Conduct the analysis in SPSS as shown on the next slide

Slide 19
 Hypothesis Tests About m 1 - m 2:
σ unknown (Assume Unequal Variances)

SPSS Output

Slide 20
 Hypothesis Tests About m 1 - m 2:
σ unknown (Assume Equal variances)

4. Compute the p –value.

For t = 2.393, the corresponding p-value is 0.037

5. Determine whether to reject H0.

Since the p–value = 0.037 is less than a = 0.05,

we reject H0.

There is sufficient evidence to support the claim that the

average carbohydrate intake by obese pregnant women is
significantly different from that of non-obese pregnant
women.
Slide 21
 Inference About the Difference Between the
Means of Two Populations: Paired Samples

 Hypotheses
We focus on tests involving difference between two
samples paired data (matched or dependent sample)

H0: p1 - p2 < 0 H0: mD = 0

H0: mD ≥ 0 H0: mD ≤ 0
Ha: mD < 0 HaH
: p:1 -mp2>>00 Ha: mD ≠ 0
a D

Left-tailed Right-tailed Two-tailed

where mD is the expected mean of the difference of the paired data

© 2014 Cengage Learning. All Rights Reserved. May not be scanned, copied
Slide 22
or duplicated, or posted to a publicly accessible website, in whole or in part.
 Inference About the Difference Between the
Means of Two Populations: Paired Samples
Let d represent the difference between the two paired sample

d - mD
 Test Statistic t
sd n
 where

d 
 di sd 
 i
( d - d ) 2

n n -1

The test statistic has follows the t distribution with n - 1 degrees of

freedom

© 2014 Cengage Learning. All Rights Reserved. May not be scanned, copied
Slide 23
or duplicated, or posted to a publicly accessible website, in whole or in part.
 Inference About the Difference Between the
Means of Two Populations: Paired Samples
 Example: Cholesterol
A dietitian wishes to see of a person’s cholesterol
level will change if the diet is supplemented by a
certain mineral. Ten subjects were pretested , and
then they took the mineral supplement for a 6-week
period. The results are shown in the table on the next
slide (cholesterol level is measured in milligram per
deciliter). Can it be conclude that the cholesterol level
has been changed at 10% significance level.

© 2014 Cengage Learning. All Rights Reserved. May not be scanned, copied
Slide 24
or duplicated, or posted to a publicly accessible website, in whole or in part.
 Inference About the Difference Between the
Means of Two Populations: Paired Samples

Cholesterol level (milligrams per deciliter)

Subject Before After
1 32 25
2 30 24
3 19 15
4 16 15
5 15 13
6 18 15
7 14 15
8 10 8
9 7 9
10 16 11
© 2014 Cengage Learning. All Rights Reserved. May not be scanned, copied
Slide 25
or duplicated, or posted to a publicly accessible website, in whole or in part.
 Inference About the Difference Between the
Means of Two Populations: Paired Samples

H0: mD = 0
1. Develop the hypotheses.
Ha: mD ≠ 0

2. Specify the level of significance. a = 0.10

3. Compute the value of the test statistic.

Conduct the analysis in SPSS as shown on the next slide

© 2014 Cengage Learning. All Rights Reserved. May not be scanned, copied
Slide 26
or duplicated, or posted to a publicly accessible website, in whole or in part.
 Hypothesis Tests About
Proportion of two population

SPSS Output

© 2014 Cengage Learning. All Rights Reserved. May not be scanned, copied
Slide 27
or duplicated, or posted to a publicly accessible website, in whole or in part.
 Hypothesis Tests About
Proportion of two population

4. Compute the p –value.

For t = 1.375, the corresponding p-value is 0.203

5. Determine whether to reject H0.

Since the p–value = 0.203 is greater than a = 0.10,

we fail to reject H0.

There is no sufficient evidence to conclude that the

cholesterol level has changed after supplementing the diet

© 2014 Cengage Learning. All Rights Reserved. May not be scanned, copied
Slide 28
or duplicated, or posted to a publicly accessible website, in whole or in part.
 Hypothesis Tests about
Proportion of two population

 Hypotheses
We focus on tests involving no difference between
the two population proportions (i.e. p1 = p2)

H 0 : p1 - p2  0 H
H00:: pp1 -
- pp2 < 00 H 0 : p1 - p2  0
1 2
H a : p1 - p2  0 H
Ha:: pp1 -
- pp2 > 00 H a : p1 - p2  0
a 1 2

Left-tailed Right-tailed Two-tailed

© 2014 Cengage Learning. All Rights Reserved. May not be scanned, copied
Slide 29
or duplicated, or posted to a publicly accessible website, in whole or in part.
 Hypothesis Tests about p1 - p2

 Test Statistic ( p1 - p2 )
z
 1 1 
p(1 - p )   
n
 1 n2 

 Pooled Estimator of 𝑝

n1 p1  n2 p2
p
n1  n2

This test statistic has a standard normal probability Distribution z

© 2014 Cengage Learning. All Rights Reserved. May not be scanned, copied
Slide 30
or duplicated, or posted to a publicly accessible website, in whole or in part.
 Hypothesis Tests About
Proportion of two population
Example: Calcium intake
A nutritionist claims that the proportion of females that
consume the WHO recommended daily intake of calcium is
different from that of the males. A recent study determined
the daily calcium intake of 20 random samples involving
males and females who visited a particular hospital and the
data is shown below on the next slide. Conduct a
hypothesis test with a = 0.05 to test the claim

© 2014 Cengage Learning. All Rights Reserved. May not be scanned, copied
Slide 31
or duplicated, or posted to a publicly accessible website, in whole or in part.
 Hypothesis Tests About
Proportion of two population
Calcium Gender
no male
yes male
no male
no male
no male
yes male
no male
no male
no male
yes male
no female
yes female
no female
yes female
no female
no female
yes female
no female
yes female
yes female Slide 32
 Hypothesis Tests About
Proportion of two population

H0: PM = PF
1. Develop the hypotheses. Ha: PM ≠ PF

2. Specify the level of significance. a = 0.05

3. Compute the value of the test statistic.

Conduct the analysis in SPSS as shown on the next slide

© 2014 Cengage Learning. All Rights Reserved. May not be scanned, copied
Slide 33
or duplicated, or posted to a publicly accessible website, in whole or in part.
 Hypothesis Tests About
Proportion of two population
SPSS Output

© 2014 Cengage Learning. All Rights Reserved. May not be scanned, copied
Slide 34
or duplicated, or posted to a publicly accessible website, in whole or in part.
 Hypothesis Tests About
Proportion of two population

4. Compute the p –value.

For z = 0.208, the corresponding p-value is 0.648

5. Determine whether to reject H0.

Since the p–value = 0.648 is greater than a = 0.05,
we fail to reject H0.

There is no sufficient evidence to support the claim

that the proportion of females who consume the
WHO recommended daily intake of calcium is
different from that of the males
© 2014 Cengage Learning. All Rights Reserved. May not be scanned, copied
Slide 35
or duplicated, or posted to a publicly accessible website, in whole or in part.