Tests Comparing Two Means
Hypothesis testing of sample means can prove to be a confounding undertaking, as there are numerous testing procedures from which we must choose. Selecting the correct one need not be such a daunting task if we examine the source and meaning of our data.
One condition that simplifies tests of 2 means is that one general null hypothesis is used for all means comparisons. The assumption that we begin with is that the means of the two populations are equal, or simply put:
Now, onto the test procedures.