fishfin wrote:I doubt those poles are including overseas military people who ussually tend to be very social conservative / republican.
They're also a very small percentage of the voting population and are possibly factored in already into polling numbers. Pollsters do weight their samples to try to match more closely the composition of the voting public.
I was also just able to get to the website you had given (It had been inaccessible before) and noticed that it put on a whole lot more blue on the map than the NYTime's pole. I have no way of knowing which one is more accurate, but on both of them Obama has been gaining ground over the past couple of days.
Of course you can know which is more accurate; study the methodology of the New York Times polls and fivethirtyeight (available at "FAQ"), and decide which you trust more. Fivethirtyeight is based on demographic analysis of each of the states combined with a composite polling average based on the reliability and bias of all the pollsters and the recentness of the polls, with some adjustment based on national polling numbers for states that haven't been polled recently. It's a bit like consolidating all the polls together to make a kind of "meta-poll" of the state of the race.
If you don't trust this site when it says that Obama's ahead (and I don't know why you wouldn't; I know it's being presented to you as a part of a disagreement but the methodology is quite sound! fwiw they did a better job forecasting the Obama vs. Clinton primaries than most pollsters and "analysts"), would you trust
someone from Princeton?
Andrew Parsonson wrote:frenchfisher wrote:Andrew Parsonson wrote:Also, the Republicans tend to perform better on election day than exit polls suggest; mainly due to people being ashamed that they vote Republican, I would guess.
Evidence?
I know
the article isn't CNN or Time or such, but you just quoted wikipedia at me so I feel justified.
That source discusses exit polls, not polls conducted before an election; they're different.
I used Wikipedia to try to help explain to you what the margin of error in a poll means, exactly... it's kind of hard to say that Wikipedia's editors would inject their own biases into an article about a statistical phenomenon, of all things.
However, let me try again. Here's a quote from
this article:
Unfortunately for the readers of this story, it is wrong. There is no reasonable statistical basis for claiming that Clinton's lead over Dole has slipped.
Why? The margin of error. In this case, the CNN et al. poll had a four percent margin of error. That means that if you asked a question from this poll 100 times, 95 of those times the percentage of people giving a particular answer would be within 4 points of the percentage who gave that same answer in this poll.
(WARNING: Math Geek Stuff!)
Why 95 times out of 100? In reality, the margin of error is what statisticians call a confidence interval. The math behind it is much like the math behind the standard deviation. So you can think of the margin of error at the 95 percent confidence interval as being equal to two standard deviations in your polling sample. Occasionally you will see surveys with a 99 percent confidence interval, which would correspond to 3 standard deviations and a much larger margin of error.
(End of Math Geek Stuff!)
So let's look at this particular week's poll as a repeat of the previous week's (which it was). The percentage of people who say they support Clinton is within 4 points of the percentage who said they supported Clinton the previous week (54 percent this week to 57 last week). Same goes for Dole. So statistically, there is no change from the previous week's poll. Dole has made up no measurable ground on Clinton.
And reporting anything different is misleading.
Don't overlook that fact that the margin of error is a 95 percent confidence interval, either. That means that for every 20 times you repeat this poll, statistics say that one time you'll get an answer that is completely off the wall.
There's also a pamphlet from the
University of Idaho if you're one of those who's suspicious of any non-.org and non-.edu site.
For two, that still means the leads is outside the margin of error.
Yes, it suggests the poll lead could be as little as .5%, which could be very easily reversed.
What I'm trying to say is: yes, you're right, the lead
could be .5%, but is in all probability
not. The confidence interval built into the margin of error emphatically
does not say that the true mean of the proportion of the population that supports Obama has an equal probability of being somewhere within the margin of error; it's a bell curve. The probability of the true mean being at 0.5% is the same as the probability being at 12.5%. In fact, they're both quite minimal, because that 6.5% number isn't from just one poll, but rather a composite of several, making the effective sample size something closer to 20,000.
For reference, I support Obama-Biden (or more specifically, strongly oppose McCain-Palin). McCain promises a further decline in the opinion of the US worldwide, and we all know how strong that is at the moment...
For reference, I've only thought Obama would be victorious
for about three weeks 
My prediction on 7/22 was an exception.
