When two groups are compared, the statistic that is used very often the t-test. It is an inferential statistic. There are a few basic facts about t-tests:
1. A t-test is used if there are only two groups to compare.
2. This statistical technique answers the null hypothesis: There is no difference between two groups on their respective mean scores.
3. There is one Independent variable with two categories and there is one Dependent variable.
Basically, the independent variable for a t-test is nominally scaled. There are two discrete categories with one variable. With a t-test there is also only one dependent variable. It is continuous in its numeric range and uses interval or ratio measurement scales.
There are two forms of the t-test. One is called the Independent samples t-test and the other is called the Correlated samples t-test, also referred to as a Paired or Matched Samples t-test.
> For the Independent samples t-test, the two groups have no relationship to each other. They are independent of one another, hence the name Independent Samples t-test.
> Correlated samples t-tests use two groups that have a connection or relationship to each other.
1. When you have two sets of scores on the same individual.
2. When you have data from sets of twins.
3. If you have matched your sample on some other variable so that the two groups are alike.
Return from t-test to statistical tests.