Question

What is Type I error? How do we correct for Type I error? What happens when we correct for Type I error? What is Type II error? How do we correct for Type II error? What happens when we correct for Type II error? How can we correct for both Type I and Type II error at the same time? Which error is considered the worst type of error to commit?

Answer 1

• Type I error is defined as the probability of rejecting the null hypothesis given the null hypothesis is false. As we fix the type I error at the start of the test which is equal to the level of significance, therefore we can lower the level of significance to correct or lower the type I error probability here. When we do this, the rejection of null hypothesis becomes difficult but this leads to the increase in probability of not rejecting a false null hypothesis as well which means that the type II error increases if we decrease the type I error.
• Type II error is defined as the probability of not rejecting a false null hypothesis. Again, as we increase the level of significance of the test, the type II error probability decreases but again this leads to an increase in the type I error probability. Therefore there is a trade off between Type I and Type II error probabilities.
• Both type I error and Type II errors can decrease at the same time by increasing the sample size for the test. The error which is considered worst type error to commit depends on the business problem for example if there is a test for some disease, then we would want to minimize the False positives instead of minimizing the false positives.