Statistics: They Can't Be Serious (Part 2)

I've mentioned the stupidity of surveys and statistics before, in Statistics: They Can't Be Serious. Today they Guardian has another one (again relating to tennis):

Go here to read it online or page 17 of the printed version.

What bugs me in particular is encapsulated in this sentence: "Swiss neuro-scientists found regular players were better at spotting subtle differences in speed, location and trajectory than others."

Isn't that the same as saying: "Regular swimmers can swim better than those who don't swim regularly?" Or: "People who read books regularly read better than those who don't?"

Or: "Stupid people who regularly put together silly statistics come up with dumber statistics than those who don't and aren't?"


Hehe it reminds me of all the other fancy studies made, one moment coffee is healthy the next it's dangerous, this goes on with diet coke, wine, salmon etc. too. I basically just laugh at these things by now and attribute it to the fact that scientists/researchers have to publish material occasionally to keep their positions. That explanation probably covers this elusterous study as well.

Posted by Casper Bang on June 10, 2008 at 09:11 PM PDT #


Posted by guest on June 10, 2008 at 11:21 PM PDT #

I guess you could be skeptical, but 49% of all researchers are above average at analyzing research results.

Another interesting statistic is that 74% of all statistics are made up on the spot. Like this one.

Posted by Anatole Wilson on June 11, 2008 at 01:00 AM PDT #

> 49% of all researchers are above average

eek - I hope that was tongue-in-cheek, otherwise I've been calculating averages the wrong way :)

As for the tennis player study, I heartily agree with you Geertjan that there are a lot of tautological and/or silly studies reported (see Journal of Improbable Research, but I don't think that applies in this case. The researchers' findings don't just mean "regular tennis players are better at playing tennis" but that there is a measurable difference in perceptual skills between regular players and others. If you follow up the original report (which I haven't so I'm going out on a limb here :) you'll find that the researchers 'controlled' for other influences on perceptual ability so that they could make inferences about the effect of playing tennis. The results could be interpreted two ways (which aren't mutually exclusive): (1) only those with superior perceptual skills play tennis regularly (2) playing tennis regularly improves your brain's ability to deal with dynamic perceptual tasks. Number (2) is obviously the most interesting.

As for coffee, I choose to ignore any studies that say it's anything less than essential :)


Posted by Michael on June 11, 2008 at 03:09 PM PDT #

Another fun stat for you: 75% of the population do not trust politicians. 25% of the population....are politicians.


Posted by Freddy D. on June 12, 2008 at 08:04 AM PDT #

Post a Comment:
  • HTML Syntax: NOT allowed

Geertjan Wielenga (@geertjanw) is a Principal Product Manager in the Oracle Developer Tools group living & working in Amsterdam. He is a Java technology enthusiast, evangelist, trainer, speaker, and writer. He blogs here daily.

The focus of this blog is mostly on NetBeans (a development tool primarily for Java programmers), with an occasional reference to NetBeans, and sometimes diverging to topics relating to NetBeans. And then there are days when NetBeans is mentioned, just for a change.


« July 2016