Hypothesis testing is done to reject or fail-to-reject a hypothesis.
One hypothesis is not accepted directly, because we can't test for all data in the world.
Steps for Hypothesis Testing
- Define a hypothesis from a sample dataset
- Do more testing on more different dataset
- if they all are opposite, then reject the hypothesis
- If they are similar, then fail-to-reject the hypothesis
If the hypothesis is "there is no difference between A & B", then it is called Null Hypothesis