Est. May 2008

29 July, 2013

Lies, damned lies, and statistics

Want a quick lesson on how to manipulate statistics?  Here ya go.

Christian Post has an article by Napp Nazworth which claims that ‘a majority would support gay marriage in all 50 states’.

The claim is made by Gallup, which conducted a poll and discovered that 'In U.S., 52% Back Law to Legalize Gay Marriage in 50 States'.  But when you go take a look at Gallup’s methodology, you come across this:
Results for this Gallup poll are based on telephone interviews conducted July 10-14, 2013, with a random sample of 2,027 adults, aged 18 and older, living in all 50 U.S. states and the District of Columbia.
And here is why Gallup’s conclusions aren’t worth the paper they’re written on (or, in this case, the pixels they used to put it on-line):

2,027 people polled from all 50 states and DC, right?

316,000,000 people in the US (estimated from the Census Department’s Population clock). 

So, you divide 2,027 by 316,000,000, then multiply by 100 to get the percentage of people in the US who were polled.

That turns out to be right around … 0.00064% of the US population.

So, 52% of 0.00064% of the US population is favorable to homosexual marriage; what's the significance?

If you figure that one out, let the rest of us know.  Far as I can tell, there is none.

No comments: