Tuesday Apr 15, 2008

Statistics: They Can't Be Serious

I'm always sceptical about statistics, on many different levels. But, principally, my issue with them is not so much in the area of gathering (although there are MANY questions to be asked there too), but in the area of interpretation. Now, I'm not the first person to suggest that interpretation renders statistics meaningless, but it's not often that the evidence is as strong as it is with the latest set of statistics I've read about. Here's the interpretation of a set of statistics, as reported in today's Guardian:

Click here to read the whole story. However, in today's Dutch Volkskrant (as reported here), commenting on the same set of statistics, the article is headlined "John McEnroe Was Usually Wrong":

The evidence here is clear. These reporters had exactly the same statistics. But they interpreted them in completely opposite directions. On top of that, one even wonders if they read the same statistics because in the Dutch report, the line judges were right in 61% of cases, while in the Guardian the line judges were wrong in 40% of them. But... that's the same thing!!! Yet... interpreted in such a way that completely different meanings are rendered. To the Guardian, the statistics support John McEnroe's objections; to the Volkskrant, the statistics undermine them.

If there's one thing that these statistics prove, it is that it pays to be multilingual and it pays to read more than one newspaper. Especially when it comes to statistics. Still, it would be nice to know whether John McEnroe was mostly right or mostly wrong. At the time, I thought he was mostly wrong, but only because I was 12 at the time and assumed that authority has special access to truth. Since then, I've learned that that's really a pretty bad assumption. How cool would it be if history were to prove that John McEnroe had been right all along? Even though he probably never thought so himself and merely enjoyed throwing tantrums on TV?


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.


« April 2008 »