In: Statistics and Probability
Describe two statistical tests which could be used for group comparison; and when each test would be appropriate
Solution:
There are two statistical tests which could be used for group comparison. These two tests are given as independent samples t test and dependent or related samples t test.
Independent samples t test:
When we need to test a hypothesis regarding the significant difference between population means of two independent populations, then we use independent samples t test. In this test scenario, two samples are completely independent and not related any way. The sample sizes of the two groups or samples may be same or different. If we know that the population variances are equal, then we use pooled variance for the calculation of test statistic. If we know the values for the population variances or standard deviations, then we use two samples z test instead of this test.
Related samples t test or paired t-test:
When we test a hypothesis regarding the significant difference between population means of two related populations, then we use paired t-test. In this test, two samples are related to each other. There is a one-to-one relationship between the observations of two samples. Therefore, both samples have same sample size.