The t test (II) concerns itself with two groups. These groups are independent; that is, an individual in one group cannot be in any way related to an individual in the other group. The purpose of the t (II), or t ratio as it is called, is to determine whether the mean of one group is significantly different from the mean of the other group. The requirements of the t (II) are as follows :
The two groups are independent.
Measurement is at least at the interval level.
The populations are both normally distributed.
The populations have the same variances. (Older statistics books and articles suggest that a test of homogeneity of variance be carried out before it is permissible to use a t ratio. However, many modern statisticians feel that such a test is not worth the time and effort. Where these tests for homogeneity of variance are most needed ” on small samples with unequal sizes ” they are the least effective [Hays, 1973].)
The samples are drawn at random.