## Alpha Levels

**Overview:**

We can define a “rare event” in a hypothesis test by setting a limit for our P-value. If the P-value gets below the limit then you reject the null hypothesis. If it’s above that alpha level then we fail to reject the null hypothesis.

**Alpha levels**

Alpha levels are labeled with the Greek letter α. The most common alpha levels to use are 0.10, 0.05, and 0.01. It’s always best to consider the right alpha level for the right situation. The alpha level is also known as the significance level. If we reject the null hypothesis we say the test is significant at that level.

**Type I and II Errors**

- Type I- The null hypothesis is true, but we reject it
- Type 2- The null hypothesis is false, but we fail to reject it

These two are easy to get mixed up so just keep in mind that we always start by assuming the null hypothesis is true, so it is the first error we can make.

**Final Reminder:**

If you’re assessing something like the safety of air bags, you would want a low alpha level. When choosing an alpha level depending on situations, for something as serious as safety you want to be as accurate as possible so you would have a low alpha level. Say you’re wondering whether your friends prefer pizza with or without pepperoni. Then you could use a higher alpha level because it’s not as serious.

**Example:**

Andre preformed a hypothesis test to test if the crime rate was higher in the city vs. lower. He arrived with a p-value of 0.009. State a valid alpha level and conclusion.

Answer: In this case due to a serious situation you would want to be pretty sure about what you’re testing, therefor you would choose a lower alpha level. In this case an alpha level of 0.01 would work seeing s the p-value is closer. Now with a p-value of 0.009 you would reject the null.