There has just been a well publicised debate between Tony Blair and Christopher Hitchens on the role of God in the world – the motion was – “Be it resolved, religion is a force for good in the world”. It was a bit of a bizarre one. Choosing one of the most controversial recent converts to Catholicism and an atheist dying of cancer. The Guardian and the Independent have called the debate for Christopher Hitchens.
But here is the issue. Prior to the debate 57% of the audience were ALREADY convinced of Christopher Hitchens views against 21% for Tony Blair. That is nearly a 3:1 margin with only 22% undecided. Yet in Canada at the census in 2001 only 16.1% said that they had no religious faith. The group with the largest atheist or agnostic grouping was in the 15-29 age group.
The Independent and the Guardian called it for Christopher Hitchens because for the Independent the final ratio was 2:1 (and no talk of the pre debate bias) and the Guardian were slightly more sophisticated and said that the undecided broke for Christopher Hitchens 11:10.
Now a few things here. If you have a crowd 3:1 in favour of someone isn’t it very likely that you will be swayed by the crowd? Isn’t it likely that the louder claps and the horrible situation that Christopher Hitchens faces might sway you? As the Guardian stated:
Throughout the 90-minute debate Hitchens seemed to have the crowd’s sympathy. That might have been to do with his ill appearance due to cancer, but was far more likely to be down to the sharpness of his verbal barbs and the fact that 57% of the audience already agreed with his sceptical position according to a pre-debate poll, while just 22% agreed with Blair’s side. The rest were undecided.
But if the ratio is 57:22 to start with and ends up with 68:32 the ratio was going in Blair’s favour (to translate those ratios – in the first batch 72% who professed a preference were for Hitchens dropping to 68% in the second).
The reality in the world is that atheists make up a very small proportion of the population of the world but around the “sacred” persons of “saints” such as Hitchens and Dawkins (the pair whom Professor Terry Eagleton rather amusingly refers to as Ditchkins) people tend to cluster. Our media is also incredibly sympathetic towards them and Dawkins gets away with things on the media that others couldn’t. Therefore it is likely (though not certain) that more people turned up at the lecture because of Hitchens. All such groups are self-selecting – including if it was one where the overwhelming number were predisposed to a religious viewpoint. Statistically the numbers have no merit to them – but of course it is easier to use them and make a story.
The reality is that people of faith contribute more to society than those that don’t have faith. As I have argued in an earlier post Indeed there was an article in USA today recently that made the same point. So, demonstrably Hitchens is wrong. The thing that I suspect went down well was the no holds barred approach he took and the fact that it is easier to score cheap points when you take this approach.
So beware of these statistical manipulations – from either side. I am a firm believer in good statistical approaches to things. The statistics for this event go no wider than the people inside the event.
One other thing was the statistics that were displayed for a world-wide survey where we are told that 48% see religion is a good thing and 52% is a bad thing. We are told that 18,192 people were surveyed – presumably to make us impressed. A few comments as I haven’t seen the full details. 8 of the 23 countries are European countries – yet they make up only 830 million of the world population of over 6 billion. So if they have a 1/3 weighting in the survey this is a nearly tripling of their true weighting. There is no China (larger than the entirety of the whole of europe) in the survey. As I have said I haven’t seen the underlying weighting but I can’t see that this is a worthwhile properly weighted survey.
Lets use statistics properly and not in this sort of way.