Two-sample t-tests

In this app you can generate two samples of random numbers and perform a two-sample t-test. There are two choices for generating the sample data-sets, either normally distributed or uniformly distributed. Remember that although you can do a t-test on any values, one of the assumptions is that the data are normally distributed, or when it is a large sample, such that the central limit theorem applies.

The null hypothesis in the two-sample t-test is that the difference between means is 0.

When you click on the check-box, a two-sample t-test will be performed. The test statistic will appear on the appropriate t-distribution and the regions beyond the critical value will be highlighted in blue. The output from the t-test will also be shown.