I'm not having a pop at the fine folks of Edelman who bring us this research each year (many of whom I have a great deal of time and respect for). I am calling on everyone to take a moment to think before they quote, tweet or otherwise pontificate based on the results.
For instance (apologies Keith, I've taken you at random from the #Trust2011 stream on twitter today (January 25, 2011)
This neither. Not with anything like a useful degree of certainty.
The danger in both cases (and they aren't alone, and I've almost certainly been guilty of doing the same myself in the past...) is that we conclude that what is true of a small subset is true for the whole populace.
You may be able to make the case if the sample was relatively representative of the general populace. Random even. But the Edelman data isn't.
And it very clearly isn't. All you have to do is read slide two of their own presentation on the Trust Barometer to know this:
Please treat it as such.
Doesn't make for as neat a headline, soundbite or tweet does it? The truth's a bitch sometimes, ain't it?
Full report etc here.
No comments:
Post a Comment