'The caveat...' links to 'Eyetracking the news', a very interesting study of how people read newspapers. The study took 582 people in the US and asked them to read newspapers and news websites while wearing eyetracking equipment to see just how much of a story was read, and what gets looked at more often on the page.
The study showed that on average, tabloid readers only read around 49% of a story. Online readers read an average of 62% of each story, but only broadsheet websites were used in the study.
As I read this post, I was reminded of 'Evidence based smear campaigns' from this May, which linked to 'When Corrections Fail: The Persistence of Political Misperceptions'.
'When Corrections Fail...' demonstrated that when people have deeply entrenched views, attempts to disprove those views can actually end up strengthening them even further.
So, in the tabloids we have stories where evidence that could disprove the whole article is buried where hardly anyone will read, and even if people do reach that far they could be likely to become more rather than less convinced of the story's truth.
As Ben Goldacre says about the evidence burials:
Caveats in paragraph 19 are common. This evidence strongly suggests that they are also a sop: they permit a defense against criticism, through the strictest, most rigorous analysis of a piece. But if your interest is informing a reader, they are plainly misleading.Quite.
Add to that the technique of putting the caveat in someone else's words - an unnamed quote from a tabloid bogeyman like a council worker is favourite - after definite statements in the newspaper's voice and named quotes from people agreeing with the paper's take that directly contradicts the caveat and you have a great recipe for misleading a whole bunch of people.
For more about the first Bad Science post, check out Angry Mob here.