When most people hear the word average they think normal, typical, boring. But, in many cases that is very far from the truth.
Averages can not only be interesting, but also impossible!
Consider the call center that measures average response time. The average response time is less than their target, so they must be doing pretty good, right???
Wrong!
When an average is less than a target value, without additional information, you have almost no idea how you are doing. You know that at least one value is less than the target, but it could be that only one value is less than the target, or it could be that all of the values are less than the target. So, our imagined call center’s comparison of their actual response time to their target value actually yields very little information. A much better approach would be if they performed a capability analysis on their response times.
Consider the following two data sets, the data sets are clearly very different, but the averages are the same.
click to enlarge
Finally, when are averages impossible? Consider the case of the average family with 1.86 children.
Before you fall into the trap of thinking averages are boring, think again.
