This tutorial talks about Type I and Type II errors. Now, since hypothesis tests are about considering the probability of a result from a sample that differs in a statistically significant way from the assumed null value, there can be errors in the outcome of the hypothesis test.
One type of error is called a Type I error. That's when the null hypothesis is rejected, even though it is true. Another type of error is a Type II error, and that's when the null hypothesis is not rejected, even though it is not true.
Now, depending on the situation you're in, we can better deal with one type of error than the other. Let's walk through an example.
Now going to look at an example of the criminal justice system. In the criminal justice system, the null hypothesis is that people are innocent. If that null hypothesis is rejected, people go to jail. If that null hypothesis is accepted, people are freed.
Now, if the null hypothesis is false, then the person is guilty. They committed the crime. So they go to jail, and that's OK. That's the result that we're looking for.
If the null hypothesis is true and it's accepted and the people go free, we're also looking for that, so that one's good, too.
Now, here's where the errors come in. If an innocent person-- so the null hypothesis is true-- if that null hypothesis is rejected, they go to jail, and that is a Type I error. On the other hand, if the null hypothesis is false and the person is freed, then that is a Type II error.
And again, depending on the situation, it's easier to correct a Type I than a Type II error. And with the criminal justice system, it's really difficult to tell which one is easier to correct.
This has been your tutorial on Type I and Type II errors.
