Two Sample Independent Test

Compares the means of two independent groups in order to determine

whether there is statistical evidence that the associated population means are
significantly different. Can be run on sample data from a normally distributed
numerical outcome variable to determine if its mean differs across two independent
group. The Independent Sample T-test is a parametric test.

Hypothesis:
𝐻𝑜: The population mean of one group equals the population mean of the
other group or 𝜇1 = 𝜇2
𝐻𝑎: The population mean of one group does not equal the population mean of
the other group or 𝜇1 ≠ 𝜇2

Relevant Equations:
Degrees of Freedom: varies by conditions, but the basic rule of thumb for a
hand circulation is the smaller of 𝑛1 − 1 and 𝑛1 − 2 where 𝑛 is the sample size for
each group.
𝑥̅1 − 𝑥̅2
𝑡=
𝑆𝐸

Definition of Two Sample Independent T-test

Test that significant difference of two independent population means.
7 steps of Two Sample Independent T-test:
1. Define the Null and Alternative Hypothesis
2. State Alpha
3. Critical Region
4. Test-Statistic
5. Computation
6. Decision
7. Conclusion

𝐻𝑜: 𝜇1 −𝜇2 = 𝑑𝑜
𝐻𝑎: 𝜇1 − 𝜇2 ≠ 𝑑𝑜 𝑜𝑟 𝜇1 − 𝜇2 > 𝑑𝑜 𝑜𝑟 𝜇1 − 𝜇2 < 𝑑𝑜
 𝑯𝒐 𝑯𝒂 Critical Region Test Statistic
𝜇1 −𝜇2 𝜇1 − 𝜇2 ≠ 𝑑𝑜 𝑡 < −𝑡𝛼⁄2,𝑣 or (𝑥̅1 − 𝑥̅2 − 𝑑𝑜)
𝑡=
= 𝑑𝑜 1 1
𝑡 > 𝑡𝛼⁄2,𝑣 𝑆𝑝√𝑛 + 𝑛
1 2

𝜇1 − 𝜇2 > 𝑑𝑜 𝑡 > 𝑡𝛼,𝑣

(𝑛1 − 1)𝑆 2 + (𝑛2 − 1)𝑆 2
𝑆𝑝 = √
𝑛1 + 𝑛2 − 2

𝜇1 − 𝜇2 < 𝑑𝑜 𝑡 > −𝑡𝛼,𝑣 𝑣 = 𝑛1 + 𝑛2 − 2

Example 1:
A statistic teacher wants to compare his two classes to see if they performed
any differently on the test he gave that semester. Class A had 25 students with an
averagescore of 70, standard deviation 15. Class B had 20 students with an average
score of 74, standard deviation of 25. Using Alpha 0.05, did these two classes perform
differently on test?
Solution:

Class A Class B 𝑣 = 𝑛1 + 𝑛2 − 2 𝛼⁄ = 0.05⁄

2 2
𝑛𝐴 = 25 𝑛𝐵 = 20 = 25 + 20 − 2 = 0.05
𝑥𝐴 = 70 𝑥𝐵 = 74 = 43
𝑆𝐴 = 15 𝑆𝑏 = 25

1. 𝐻𝑜: 𝜇𝐴 − 𝜇𝐵 = 0
𝐻𝑎: 𝜇𝐴 − 𝜇𝐵 ≠ 0

2. 𝛼 = 0.05

3. Critical Region:
Reject 𝐻𝑜 if 𝑡 < −2.017, 𝑡 > 2.017

4. Test-Statistic:
 Two Sample Independent T-test

5. Computation:

(𝑛1 − 1)𝑆1 2 + (𝑛2 − 1)𝑆2 2

𝑆𝑝 = √
𝑛1 + 𝑛2 − 2

(25 − 1)(15)2 + (20 − 1)(25)2

=√
25 + 20 − 2

= 20.0436

(𝑥̅1 − 𝑥̅2 − 𝑑𝑜)

𝑡=
1 1
𝑆𝑝√𝑛 + 𝑛
1 2

(70 − 74 − 0)
=
1 1
20.0436√ + 20
25
= −0.6652
6. Decision:
Accept 𝐻𝑜 since −0.6652 > −2.017

7. Conclusion:
Therefore, there is no significant difference between the test
Performance of Class A and Class B, 𝑡 = −0.67, 𝑝 > 0.05.

Example 2:
A researcher at a University wants to determine if female college students
sleeps significantly longer than male college students. They collect data on a random
sample of 51 female college students and find they sleep an average of 7.95 hours
(standard deviation of 1.88 hours) per night, while 41 male college students report
sleeping an average of 7 hours (standard deviation of 3.06 hours) per night. Conduct
the appropriate test to determine if female college students sleep significantly longer
than male college students assuming that the population variances are not equal.
Solution:
1. 𝐻𝑜: 𝜇𝐴 − 𝜇𝐵 = 0
𝐻𝑎: 𝜇𝐴 − 𝜇𝐵 > 0

2. 𝛼 = 0.05

3. Critical Region:
Reject 𝐻𝑜 if 𝑡 < −1.684, 𝑡 > 1.684

4. Test-Statistic:
Two Sample Independent T-test

5. Computation:

(𝑛1 − 1)𝑆1 2 + (𝑛2 − 1)𝑆2 2

𝑆𝑝 = √
𝑛1 + 𝑛2 − 2

(51−1)(1.88)2 +(41−1)(3.06)2
=√ 51+41−2

= 2.4749

(𝑥̅1 − 𝑥̅2 − 𝑑𝑜)

𝑡=
1 1
𝑆𝑝√𝑛 + 𝑛
1 2

(7.95 − 7) − 0
=
1 1
2.4749√ +
51 41
= 1.8299

6. Decision:
Reject 𝐻𝑜 since 1.8299 > 1.684

7. Conclusion:
Therefore, female college students do sleep longer than male college
students.
Exercises:
1. An investigator thinks that the people under the age of forty have vocabularies
that are different than those of people over sixty years of age. The investigator
administers a vocabulary test to a group of 31 younger subjects and to a group of
31 older subjects. Higher scores reflect better performance. The mean score for
younger subjects was 14.0 and the standard deviation of younger subject’s scores
was 5.0. The mean score for older subjects was 20.0 and the standard deviation of
older subjects was 6.0. Does this experiment provide evidence to the investigators
theory?
a) Provide the null and alternative hypothesis.
b) Using an alpha level of 0.05
c) State your conclusion regarding the results from this test.

Exercise 2:
An investigator predicts that dog owners in the country spend more time
walking their dogs than do dog owners in the city. The investigator gets a sample of
21 country owners and 23 city owners. The mean number of hours per week that city
owners spend walking their dogs is 10.0. The standard deviation of hours spent
walking the dog by city owners is 3.0. The mean number of hours of the country
owners spent walking their dogs per week was 15.0. The standard deviation of the
number of hours spent on walking the dog by the owners in the country was 4.0. Do
dog owners in the country spend more time walking their dogs than the dog owners
in the city?
a) Provide the null and alternative hypothesis.
b) Using an alpha level of 0.05
c) State your conclusion regarding the results from this test.