Module 2

Test of Hypothesis: Two-

Sample t-test (Equal
Variance Assumed)
 OBJECTIVES:
At the end of this module, you should be
able to:

1. Perform two-sample t – test with

assumed equal variances.
Review:
STEPS IN TESTING HYPOTHESIS
(Critical Value Method)
1.Determine Null Hypothesis 𝑯𝒐 and Alternative
Hypothesis 𝑯𝒂 .

2. Identify the Statistical Test to be used, the value of

∝, and the critical value of the test statistics.

3. Computation

4. Decision (reject or not to reject 𝑯𝒐 )

5. Conclusion ( in non-technical terms )

Key Concepts:
1.Student t - distribution.(Quarter 1 – Module 6)

2. Critical Values and Rejection Region.

(Quarter 2 – Module 5)

3. The independent t –test.

Independent t -test
1.Difference between 2 independent sample
groups. Comparing 2 groups by evaluating the
mean difference.
How to Evaluate?
- by comparing the sample means.

2. Must have 2 independent samples and 1

dependent variable that is continuous to
compare them on.
Examples
1.Do males and females differ significantly on their level of
Math anxiety?

Independent Variable:
Gender( Male and Female)
Dependent Variable:
Level of Math anxiety

2. Do older people exercise less significantly than younger

people?

Independent Variable:
Age(young and old)
Dependent Variable:
Frequency of Getting Exercise
What are some examples
that you can come up
 Null & Alternative Hypothesis
Null Hypothesis:
There is NO significant difference between the 2 groups.
𝑯𝒐 : 𝝁𝟏 = 𝝁𝟐

Alternative Hypothesis:
There is a Statistically significant difference between the 2 groups.
Non-directional:
𝑯𝒂 : 𝝁𝟏 ≠ 𝝁𝟐
Directional:
𝑯𝒂 : 𝝁𝟏 > 𝝁𝟐
𝑯𝒂 : 𝝁𝟏 < 𝝁𝟐
 Degrees of Freedom
Degrees of Freedom = sample size of group 1 + sample size of group 2 - 2

𝒅𝒇 = 𝒏 + 𝒎 − 𝟐

* We subtract 2 since we have 2 samples.

Identifying Critical Values:
In finding for the critical value of t using the t distribution
table, use the following steps as derived in the previous t-
distribution topic (module 6, quarter 1).
a. On the t- table, locate the given degree of
freedom, df.
b. On the topmost row, locate the given area,𝜶. For
𝜶
two tailed, locate the given area divided by 2,
𝟐
c. The value corresponding to the intersection is the
critical t– value.
The t-table:
Let’s Practice:
Find the critical value.

1. One tailed to the left, level of significance is 0.05,

sample 1 has 6 samples, sample 2 has 8 samples.
2. One tailed to the right, level of significance is 5%,
sample1 and sample both have 10 samples.
3. Two tailed, level of confidence 95%, sample 1 has 8
samples, sample 2 has 10 samples.
 Test statistics:
ഥ −𝒀
𝑿 ഥ
𝑻=
𝟏 𝟏
𝑺𝒑 +
𝒏 𝒎
Where:
ഥ = mean of the first group
𝑿
ഥ = mean of the second group
𝒀
𝒔𝒑 = pooled standard deviation
𝒏 = sample size of the first group
𝒎 = sample size of the second group

For pooled standard deviation,𝒔𝒑 , it can be calculated as:

𝒏 − 𝟏 𝑺𝟏 𝟐 + (𝒎 − 𝟏)𝑺𝟐 𝟐
𝑺𝒑 =
𝒏+𝒎−𝟐
Where:
𝑺𝟏 = sample standard deviation of the first group
𝑺𝟐 = sample standard deviation of the second group
𝒏 = sample size of the first group
𝒎 = sample size of the second group
Example A manufacturer claims that the average tensile strength of thread X exceeds the average tensile strength of thread Y by at
least 10 kilograms. To test his claim, 15 pieces of each type of thread are tested under similar conditions. Type X thread had an
1: average tensile strength of 85.7 kilograms with standard deviation of 5.67 kilograms, while thread Y had an average tensile
Solution : strength of 75.3 kilograms with standard deviation of 4.46 kilograms. Assume the population to be approximately normal with
equal variances and test the manufacturer’s claim using a 0.01 level of significance.
Let Thread X be the 1st Group; Thread Y be the 2nd group
1. Null Hypothesis 𝐻𝑜 and Alternative Hypothesis 𝐻𝑎
𝐻𝑜 : 𝜇1 = 𝜇2 (tensile strength is the same)
𝐻𝑎 : 𝜇1 − 𝜇2 > 10(thread X exceeds thread y by at least 10)
2. Statistical Test = t-test (one-tailed)
𝜶 = 𝟎. 𝟎𝟏, 𝒅𝒇 = 𝒏 + 𝒎 − 𝟐 = 𝟏𝟓 + 𝟏𝟓 − 𝟐 = 𝟐𝟖, 𝒕𝒄𝒓𝒊𝒕𝒊𝒄𝒂𝒍 = 𝟐. 𝟒𝟔𝟕
3. Computation: 𝑋ത = 85.7, 𝑆1 = 5.67, 𝑛 = 15; 𝑌ത = 75.3, 𝑆2 = 4.46, 𝑚 = 15

𝒏 − 𝟏 𝑺𝟏 𝟐 + (𝒎 − 𝟏)𝑺𝟐 𝟐 15 − 1 5.67 2 + (15 − 1) 4.46 2

𝑺𝒑 = = = 5.10
𝒏+𝒎−𝟐 15 + 15 − 2

ഥ−𝒀
𝑿 ഥ − 𝟏𝟎(𝒆𝒙𝒄𝒆𝒆𝒅𝒔 𝒃𝒚 𝒂𝒕 𝒍𝒆𝒂𝒔𝒕 𝟏𝟎) 85.7 − 75.3 − 10 0.4
𝑻= = = = 𝟎. 𝟐𝟏
𝟏 𝟏 1 1 1.86
𝑺𝒑 𝒏 + 𝒎 5.10 +
15 15
4. Decision (reject or not to reject 𝐻𝑜 ):
Since the computed T is less than the critical T, 𝐻𝑜 is NOT REJECTED.
5. Conclusion:
The manufacturer’s claim is not true that the tensile strength of thread X exceeds that of thread Y by more
than 10 kilograms.
 To find out whether a new serum will arrest leukemia, 9 mice, all with an advanced stage of
disease, are selected. Survival times, in years, from the time the experiment commenced are as
follows: Treatment 2.1 5.3 1.4 4.6 0.9
No Treatment 1.9 0.5 2.8 3.1
At 0.05 level of significance, can the serum be said to be effective? Assume the 2 distributions to
be normally distributed with equal variances. σ 𝑦𝑖 1.9 + 0.5 + 2.8 + 3.1
Let Treatment be the 1st Group; No Treatment be the 2nd group 𝑦ത = =
𝑛 4
1. Null Hypothesis 𝐻𝑜 and Alternative Hypothesis 𝐻𝑎 8.3
𝑦ത = = 2.075
𝐻𝑜 : 𝜇1 = 𝜇2 (serum is same without treatment) 4
𝐻𝑎 : 𝜇1 > 𝜇2 (serum is effective, since treatment group is greater) ෍ 𝑦 2 = 1.92 + 0.52 + 2.82 + 3.12
2. Statistical Test = t-test (one-tailed)
𝜶 = 𝟎. 𝟎𝟓, 𝒅𝒇 = 𝒏 + 𝒎 − 𝟐 = 𝟓 + 𝟒 − 𝟐 = 𝟕, 𝒕𝒄𝒓𝒊𝒕𝒊𝒄𝒂𝒍 = 𝟏. 𝟖𝟗𝟓 ෍ 𝑦 2 = 21.31
3. Computation: Find the sample mean and sample standard deviation.
෍ 𝑦 = 1.9 + 0.5 + 2.8 + 3.1 = 8.3
σ 𝑥𝑖 2.1 + 5.3 + 1.4 + 4.6 + 0.9
𝑥ҧ = =
𝑛 5 𝑛 σ 𝑦2 − σ 𝑦 2 4 21.31 − (8.3)2 85.24 − 68.89
14.3 𝑠2 = = = = 1.167
𝑥ҧ = = 2.86 𝑚(𝑚 − 1) 4(3) 12
5
𝒏 − 𝟏 𝑺𝟏 𝟐 + (𝒎 − 𝟏)𝑺𝟐 𝟐 5 − 1 1.971 2 + (4 − 1) 1.167 2
෍ 𝑥2 = 2.12 + 5.32 + 1.42 + 4.62 + 0.92 𝑺𝒑 =
𝒏+𝒎−𝟐
=
5+4−2
= 1.67
ഥ−𝒀
𝑿 ഥ 2.86 − 2.075 0.785
𝑻= = = = 𝟎. 𝟕𝟎
෍ 𝑥 2 = 56.43 𝟏 𝟏 1 1 1.12
𝑺𝒑 𝒏 + 𝒎 1.67 +
5 4
෍ 𝑥 = 2.1 + 5.3 + 1.4 + 4.6 + 0.9 = 14.3 4. Decision (reject or not to reject 𝐻𝑜 ):

Since the computed T is lesser than the critical T, 𝐻𝑜 is NOT REJECTED.

𝑛 σ 𝑥2− σ𝑥 2 5 56.43 − (14.3)2 282.15 − 204.49
𝑠1 = = = = 1.971 5. Conclusion:
𝑛(𝑛 − 1) 5(4) 20
Prolonging the life of mice with leukemia is the same with or without treatment.
A certain health authority set up an investigation to examine the ages of mothers when they
give birth to their first child. A random sample of 10 first-time mothers from a certain year has a
mean age of 23.48 years with a standard deviation of 2.5 years. A further random sample of 15
first time mothers taken 10 years later had a mean age of 24.01 years with a standard deviation
of 2.4 years. Test whether or not these figures suggest that there is a difference in the mean
age of first time mothers between these two dates using a 0.01 level of significance. Assume
the population to be approximately normal with equal variances.
To find out whether a new drug wil cure diabetes, 15 mice with an advanced stage of disease
are selected. Survival times, in years, from the time the experiment commenced are as follows:
Treatment 8.3 5.8 6.8 1.9 5.9 8.8 4.7 6.1
No Treatment 1.2 2.3 2.5 5.6 2.4 3.1 4.1
At 0.01 level of significance, can the drug be said to be effective? Assume the 2 distributions to
be normally distributed with equal variances.
Teacher Ryan conducted a review class in his Statistics class. He gave a test before and after
review and gathered the following data:
STUDENT
1 2 3 4 5 6 7 8 9
Score Before Review 15 16 19 15 24 24 26 11 18
Score After Review 17 10 28 20 32 28 19 30 27

At 0.01 level of significance, is the review class effective? Assume the population to be
approximately normal with equal variances.
A researcher claims that the average life span of mice can be extended by as much as 8 months
when the calories in their food are reduced by approximately 35% from the time they are
weaned. The restricted diets are enriched to normal levels by vitamins and protein. Suppose
that a random sample of 9 mice are fed with a normal diet and live an average life span of 33.2
months with a standard deviation of 3.3 months, while a random sample of 12 mice are fed the
restricted diet and live an average life span of 38.5 months with a standard deviation 2.7
months. Test the hypothesis at 0.05 level of significance that the average life span of mice on
this restricted diet is increased by 8 months. Assume the distribution of life spans for the
regular and restricted diets are approximately normal with equal variances.
THANK YOU AND GOD BLESS!