8.6 - The P-value approach to hypothesis testing

One of the differences between estimation and testing is that tests are often designed to make a specific decision. That is, we conclude a test by either rejecting a claim or not rejecting it.

Having said this, we do not always want tests to be so black and white. Yes, there are circumstances where a choice between two options must be made. But often we would like to know the extent to which the evidence in a test pointed one way or the other. To address this, there is a slight alternative to the approach we have been using so far in this chapter. This is known as the P-value approach.

In this section we will describe the philosophy of the P-value method and explain how it differs to the more traditional method of hypothesis testing we have seen thus far.

Learning objectives

In this section you will learn about:

  1. assessing how likely or unlikely a sample is in relation to the null hypothesis of a test
  2. the P-value of a test statistic
  3. the P-value approach to hypothesis testing