Lies, damned lies and headli(n)es

Here's a great example of how deceitful headlines can be and why you should always read beyond them.

Today, the French daily Le Monde ran an article entitled, 'According to a poll, nearly a third of the world population accepts the use of torture.'

But then when you actually read the article, you find out that the survey in question only interviewed people in 25 countries including those 'most at risk of terrorist attack.'

So how can a survey of 25 countries (in a world with around 200 of them) be considered a representative sample of world population? Were there African countries in the poll? Central Asian ones? Central American ones?

Totalitarian regimes almost certainly would not allow such a poll to be taken in their countries. Such regimes are also the ones most likely to actually use torture on its citizens. And when you live in a country where a 'terrorist' is a honest journalist or anyone belonging to an opposition party (or a Quaker), that might give you a different opinion of torture.