HoodaThunk?

There are 3 types of lies, the saying goes: lies, damned lies, and statistics. A good mathematician can solve a formula and give you a great depiction of the real answer. A good statistician can take than same data and make it depict any damn thing he’d like it show. Iowahawk has a hilarious-because-it’s-true post up about that wonderous “margin of error” we’re all hearing about when political polls are reported:

Works pretty well if you’re interested in hypothetical colored balls in hypothetical giant urns, or survival rates of plants in a controlled experiment, or defects in a batch of factory products. It may even work well if you’re interested in blind cola taste tests. But what if the thing you are studying doesn’t quite fit the balls & urns template?

One of the things my stats instructor taught me was that when the subjects of the analysis are intelligent they have the ability to act in random or non-random fashions at will. That automatically skews any kind of statistical analysis because statistics rely on the given characteristic being studied to remain constant. In Iowahawk’s example, a given ball of a given color is always that color regardless of whether it’s sampled or not. Human beings’ political positions don’t behave that way.

You really need to read that whole post because it’s seriously on-target.