Let’s play “How to make statistics say what you want, by CNN”
First, here’s the headline: Nearly 1 in 3 Americans Believe Obama is a Muslim (Only 39% believe he’s a Christian).
Second, let’s do some quick math: “Nearly 1 in 3” is somewhere near 33%. So “Nearly 33% of Americans believe...”
Third, the word “Nearly” is used to guide the reader into thinking “total”. Think about the terms “Nearly complete”, “Nearly ready” etc...they implies “almost full”. So in this context, the word “Nearly” is meant to make the reader see the term “1 in 3” and think large, full.
Fourth the use of “39%”. Everyone knows that 39% is a small fraction of the whole. It stands on its own. No one looks at 39% and thinks “a lot”
So the headline author, by way of using “nearly 1 in 3” to make people think “a lot” and “39%” to make people think “a small fraction” tortures the statistics into saying what he wants you to think: They want you to walk away thinking “A large portion of people believe Obama’s a muslim and a small number of people think he’s Christian” when the statistics say the opposite.
Oh, CNN, you sly dogs. Orwell taught us how to read you. We’re onto your language games.
I noticed that also. I don’t even have to think about it anymore, it’s reflexive for me to read mainstream media headlines and stories this way.