The Internet was abuzz recently about the revelation that according to a high-school class experiment, Oreo Double Stuf cookies are, in fact, not double stuffed but only 1.86 percent stuffed. The original blog post by the math teacher was actually made on March 3, 2013, but not discovered as “news” by The Huffington Post until months later.

You have to admire the teacher for doing something engaging to get the students excited about doing math, but a further educational opportunity was missed. Namely, does a single experiment on a single group of cookies from only two bags (one original, one Double Stuf) “prove” that all Oreo Double Stuf cookies are only 1.86-percent larger than the original?

It’s both a lesson in manufacturing and in statistics. It’s a classic example of thinking that measuring a single sample or lot is sufficient to indicate process variability—and by extension to determine if a process is in or out of control.

 …