Statistics

The old saying that you can prove anything with statistics is plain wrong. What is really happening is that statistics are illustrating how complex a problem or question really is. When you apply measures and confidence to a problem, you are immediately limited to that part of the problem that you have quantified. Suddenly it becomes very clear that the problem has many, many facets and that there are complex interactions between them. The world that looked so simple unveils some of its complexity when you are required to describe it numerically. Mathematics in general is complex because it is an expression of the real world, which is far more complex that mathematics.

In college classes, statistics is presented as a tool that can be applied to most real world problems. In truth, statistics can be applied to very few real world situations. The real world is so complex and misbehaved that the equations we have at our disposal are insufficient for describing or understanding it. That is why real problems have to be decomposed into smaller parts. Our equations can describe these. But, once decomposed we can only make claims or predictions about those smaller pieces. Explaining the many combinations and limitations that emerge from this is where statistics gets its reputation for being criminally manipulable. Mathematicians and statisticians are constantly pushing the field into more complex equations in an attempt to create equations that can represent more of the world.

Statistics are not trusted by most people because they are too complex for them to understand. There is no quick and easy way to understand statistics. You have to start at the beginning and work your way through the complex maze. No matter how many USA Today surveys and graphs you read, you will never learn statistics from those. There are just some fields that are best left to the professionals – quantum mechanics, electrodynamics, statistics, and auto repair.

So academics in all fields find that they must master some level of statistics (but not auto repair). It is taught is most departments and is embraced and abhorred by everyone. Since it is so broadly used, it has to be broadly talked about and everyone has an opinion on it. I have no opinion on quantum mechanics (good, bad, or useful) because I do not ever think about it. But every academic has an opinion on statistics.