Two-sample Paired t-test Using R

The two-sample t-test is crucial, nowadays, for the majority of industries. For example, in quality control departments, the two-sample t-test is used to test the response of new experiments or treatment done on products. In this test, the number of samples must be equal because the samples are related.

Using Rcmdr, click on statistics, means and paired t-test as shown in the below figure. We need to solve an example to better explain this test. Assume we have a number of chickens that will be exposed to a certain treatment.

Ho: There is no significant difference between two samples.

H1: There is a significant difference between two samples.

Following are the weights before and after treatment. We need to perform the test to conclude, whether there is a significant difference or not between the two samples. In another way, the treatment was effective or not.

Before

After

22

21

24

32.5

19

18

16

14.5

25

24

15

14.5

18

17

14

12

26

25

The results of the test are shown in the below figure. P-value (0.9602) ≥ 0.05. Thus, we accept the null hypothesis (Ho). There is no significant difference between the two samples. Thus, the treatment has not any effect on the weight of the chickens.