In Professor Gibson’s statistics class, we have been told that the best way to understand what a problem is asking is to “state it statistically”. I couldn’t agree more. But I find it a little more interesting to understand *why* we do certain problems in the first place. A vast majority of time, the problems we analyze in class are relevant to fields that are completely beyond my interests.

For example, determining the average length of dog hair amongst European shi-tzus is not on my list of things to retain from Kogod. Nor is figuring out the likelihood that average U.S. funeral fees exceed the 2005 average of $6,500 as stated by the National Funeral Directors Association (if you’re dying to check it out, here is their website).

What I find most interesting with this class are the actions that can be made *after* statistical conclusions have been reached. Let’s say, as a recent in-class example presented to us, that we have a machine which is calibrated to fill Candy-Cane-Coated Cereal boxes with 13.0 ounces of deliciously bad-for-you Candy-Cane-Coated Cereal. The problem, which gives other necessary statistical information, asks us to determine whether or not the machine is actually filling the boxes with 13.0 ounces. Through the use of certain formulas, we can figure this out and make a statistically sound conclusion about our machine. But now what?

Let’s say for instance that we conclude that our machine is filling the cereal boxes incorrectly. This is generally where the textbook problem ends. I like to take it a step further and ask “what the heck is the manager supposed to do now?”. Answering this question usually gives me a better idea of the usefulness of statistics.

In the example, the boxes had a target weight of 13.0 ounces; and for good reason. The company determined that they would make the most profit per unit at this weight given they charge a certain price per box. They likely chose this amount because this is how much, on average, is demanded by consumers on a weekly basis since grocery shopping is generally a weekly affair. If the machine filling the boxes malfunctions and boxes start becoming under-filled, the manager risks customer complaints and apologetic actions (like sending out gift certificates) to keep his/her customers. On the other hand, if the boxes start becoming over-filled, the manager risks being inefficient with his/her materials, and giving the consumer more than he/she deserves for they price paid for the cereal. Both scenarios risk to negatively lower profits. It is up to the manager to make an adjustment to his/her machine and maintain 13.0 ounces to the best the machine’s ability.

I don’t expect to go into the cereal business either, but that doesn’t mean the skills learned from this type of problem are useless. We are learning know-how, not necessarily about other industries. Looking at statistics from this perspective, however, makes these other industries rather interesting.

*“Smoking is one of the leading causes of statistics.”**– Fletcher Knebel*

I agree with the point on the obscure examples… I also feel that too often examples are outdated and narrow. Authors seem to depend on the classic factory example in a time that internet start-ups, creative small businesses and innovative entrepreneurs are huge drivers in the business world.