What George Will should have written

We’ve avoided piling on to the George Will kerfuffle, partly because this was not a new story for us (we’d commented on very similar distortions in previous columns in 2004 and 2007), but mostly because everyone else seems to be doing a great job in pointing out the problems in his recent columns.

We are actually quite gratified that a much wider group of people than normal have been involved in calling out this latest nonsense, taking the discussion well outside the sometimes-rarefied atmosphere of the scientific blogosphere (summary of links). Maybe RealClimate has succeeded in its original aim of increasing the wider awareness of the scientific context? However, like many, we are profoundly disappointed in the reaction of the Washington Post editors and George Will himself (though the ombudsman’s column today is a step in the right direction). It would have been pleasant to see an example of the conservative punditocracy actually learning something from the real world instead of resorting to ever-more unconvincing pseudo-legalistic justifications and attacks on the messenger to avoid taking their head out of the sand. Nonetheless, in a moment of naive optimism, we have allowed ourselves to indulge in a fantasy for how a more serious columnist might have dealt with the issue:

The scientific method in journalism

Feb 29th, 2009, Washington post

This column recently reported and commented on some developments pertinent to the debate about whether global warming is occurring and what can and should be done.

It is no secret that I am a critic of sensationalism in the coverage of environmental issues and that I have a philosophical preference for reality-based policies over those based on the ideologically-based fantasies of those I critique.

In my last column, I reported on a statistic concerning sea ice extent – that global sea ice extent is unchanged since 1979 – that was trivially shown to be untrue, and for that I apologize. Rather than throw the fact checkers in my office or at the Washington Post under the bus, I take full responsibility for the mistake. However, as with good scientific practice, this provides an example of how journalism too can learn from its mistakes.

The source of the original quote was a Daily Tech blog post published in early January. While that post itself was heavily criticized as being misleading, it did use data from a reliable scientific source which was technically accurate at the time. My error was in assuming that scientific ‘facts’ don’t change over a month or two and thus it was not necessary to revisit the source of the original data before writing my column. What was true in January would still be true in February, right? Wrong.

What I didn’t consider was that in complex and noisy data there are always going to be outliers, and in heavily politicised subjects there will always be people who will want to exploit a chance occurrence for a sound-bite. I should of course have known better since I decry this practice on a regular basis in discussions of economic issues. Through a combination of wishful thinking and time constraints, my failure to recognize a piece of classic cherry-picking lay at the heart of this problem.

Page 1 of 2 | Next page