Type II Error

Lots of bad Type 2's. Diabetes. Okidata. Times New Roman.

A Type II Error happens when we fail to reject the null hypothesis in a hypothesis test when the null hypothesis is actually false.

We used to work in a candy factory. Lots of little orange dudes running around. You know the one. One day, we got word that the candy bars being sent out weighed significantly less than the 96 grams they were supposed to weigh. We conducted a hypothesis test at the 0.05 alpha-level to determine if there was sufficient evidence to suggest that the candy bars did weigh less than 96 grams.

We got a p-value of 0.18, indicating that we should not reject the null hypothesis...meaning we believe our candy bars are the correct weight after all. As such, we didn’t bother to examine the machine that made the bars to ensure everything was hunky-dory. It turns out that the candy bars were actually too small. Our decision to not examine the faulty machine (because the hypothesis test results suggested there was no reason to, a Type II Error in this case) led to us making some of our customers mad, and they subsequently stopped buying our bars.