Anti-Climacus

This week in "if you don't understand statistics, don't pretend like you do":

I make no claim to be great at statistics, though I did take two graduate-level semesters and can at least fake my way around an econometric model (hint for grad students: the problem is almost always in the theoretical model, or the lack thereof). However, I do enjoy a good exploration of statistics and how people fail to understand them. So I liked this rebuttal of the recent studies linking diet soda consumption to weight gain. The theoretical model left something to be desired, apparently:

What’s that? A correlation, you say? Why, the only possible explanation is that the variable randomly assigned to the x axis must have caused the differences in the variable plotted on the y-axis! It’s SCIENCE!...

Because there’s no chance there’s some confounding factor, or that the causal arrow points in the other direction. After all, people who are getting fatter wouldn’t have any reason to be more likely to drink diet soda, would they?

And this is to say nothing of the other obvious alternative explanation, that people who consume 'diet' beverages think this frees them up to consume extra calories elsewhere, and end up consuming far out of proportion to their 'savings.'

mgoblog also had a piece on statistics and how college football fans fail to understand them. The best part:

This is the disconnect. While what seems like a fairly large subset of the fanbase saw wholesale collapse in the Wisconsin game, computers saw two units failing immensely and an offense that put up 442 yards on a defense that gave up 321 on average, scored 31-ish points (computers will credit the offense with acquiring the field position for the field goal and deduct the miss from the special teams; if they deduct from the garbage TD they will use a lower denominator when trying to figure out expected points) on a defense that gave up 21. Statistically, Michigan's offense was at least a standard deviation above the mean against the Badgers.

While the Wisconsin game is the biggest outlier between the offense's actual and perceived performance, it's instructive. ...This is why statistics are useful, because meat-emotions often overwhelm our capacity for reason.

...followed, naturally by 100-ish comments that argue statistics are meaningless and we should just believe what our eyes tell us and hurr durr requisite link to Fire Joe Morgan RIP.

The past six months has been a useful exercise in learning that Michigan fans are not particularly more intelligent than any other sports fans, our own self-mythology notwithstanding, and there are a surprising number of people who believe that if you use and attempt to understand statistics, they will steal your soul.