17 Assumptions of the t-Test
t-tests are what are known as parametric tests. In other words, they are based on the normal distribution and, as such, they rely on meeting certain assumptions. Before you report the results of a t-test, you should always check that you meet the assumptions.
t-tests are based on the assumptions that:
- The dependent variable is interval or ratio data.
- The sampling distribution is normally distributed. For the dependent samples t-test, this means that the sampling distribution of the differences between the scores should be normally distributed.
The independent samples t-test also assumes that:
- Homogeneity of the variance: variances in the populations are equal.
- Scores in the different groups are independent (because they come from different people, they should be!).
