FBIHOP Comments
Uhm, wrong. That poll doesn't say she's ahead.
It says that if we took 100 polls, sampling the same population, asking the same question, then in 95 of those 100 polls she'd come out somewhere between 6 points ahead and 4 points behind.
Of course, in the other 5 polls she'd be outside that range, and we don't really know whether the population sampled bears any relation to the population which will actually vote.
Let me try that explanation again, because I think I botched it.
They talked to 388 people out of a total population of over a hundred thousand. That sample may, or may not, be representative of the whole population.
A MOE of 5%, with Madridahead by 1% in the small sample, means that the odds are 19 out of 20 (95%) that if they asked everyone in the whole group, all 100,000+, Madrid would be somewhere between 6 points ahead and 4 points behind.
Having a 95% chance of being somewhere between down 4 and up 6 is not at all the same as being up 1.
And, of course, that assumes that those among the 388 who were lying balance each other out.
You are correct, but I was simplifying it; with the MOE what it is and the results of the poll what it is, that means that it is more likely than not that Madrid is ahead of Wilson.
If i interpret it correctly, it is more likely to be at Madrid +1% than Wilson +4% or Madrid +6%.
Thank you for clarifying, though.
Commenting by HaloScan
