A hypothesis is a statement about the value of a population parameter
- To carry out a test you have the null hypothesis $H_0$ which you assume to be true
- The alternative hypothesis $H_1$ tells you about the parameter if $H_0$ is false
For a one-tailed test: $H_0: P=n$, $H_1: P>n$ or $H_0: P=n$, $H_1: P<n$
For two-tailed test: $H_0: P=n$, $H_1: P≠n$
If the probability of $H_0$ < significance level then you reject it. A critical region is a region of the probability distribution which if the test statistic falls in you reject $H_0$ The actual significance level is the probability of incorrectly rejecting the null hypothesis