The formula used to calculate the margin of error is on the page I linked to, 1/sqroot(sample size). The size of your population has nothing to do with this. I'm not a statistician, so I'll just defer to the 'here's a bunch of math, trust me, it works' method.
On the TV polls you will see an error margin of 3-5%, and the sample size is usually shown as around 1000, which works out according to the formula. There will always be uncertainty unless you're polling the entire population. But there are bigger fish to fry:
- how is the sample selected
- what/how question is asked
- how are the results aggregated (if multiple questions) and interpreted
In election polls a change poll-to-poll of 1%, when your error margin is 5%, is meaningless. Similarly, if the answers are within margin of error points of each other, it's a toss up.
Example poll with a 4% error margin:
Quote:
http://www.robertniles.com/stats/margin.shtml
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.
|
I wasn't saying polls are good (or bad for that matter), just pointing out that a sample size of 2515 will give you pretty good results, better than you'd expect, IF DONE PROPERLY. Statistics, bell curve and probability kinda works in a funny way like that. And the old saying holds true: you can 'prove' anything with statistics.
As a side note, if only 30% of the eligible population votes, is that an election or a poll?