Learning Objectives

When you have completed this part of the Module you should be able to

• understand the concept of hypothesis testing,
• explain it to others,
• select an appropriate hypothesis test for some common cases,
• conduct a hypothesis test,
• analyze the outcome of a hypothesis test done by you or someone else, and draw conclusions,
• explain why do we want to conduct hypothesis tests.

Definition

A statistical hypothesis is an assumption, statement or claim, which may or may not be true. Commonly we formulate two hypotheses a, so called, null hypothesis (H₀) and an alternative hypothesis (H₁). The following six step procedure can be used to conduct most common hypothesis tests.

1. H₀: State the null hypothesis
2. H₁: State the alternative hypothesis
3. Select a level of significance for the test; commonly = 0.1; 0.05; 0.01
4. Select an appropriate test statistic, and determine the critical region
5. Evaluate the test statistic from a random sample
6. State the conclusion: Do not accept the null hypothesis, H₀, if the test statistic value falls in the critical region, otherwise accept H₀.

Note: Please keep in mind that you may use the above steps in general and in most cases when conducting hypothesis testing. The procedure remains the same, only the case or application situation changes.