00:01
A pole has a margin of error of 2 .5%.
00:04
Does that mean the width of the confidence interval is 5%.
00:07
So let's go through how we get confidence intervals.
00:11
We want to estimate a parameter, in this case the population proportion.
00:18
So the percentage that maybe agree with something.
00:21
But it's not reasonable to ask every single person in the population.
00:25
For example, if you want to know what percentage of americans love cats, you can't ask all of them.
00:30
It would just take too long, it would be too expensive.
00:33
So you take a sample and use that to make an estimate.
00:36
Now, i know that if i took every sample of a given size, took the sample proportions and plotted them out, i would see something approximately normal.
00:47
This is a normal approximation to the binomial, because you have n independent files, members of your sample, two outcomes for each, they love cats or they don't, same probability p, the population proportion, but they do.
01:00
The mean of this curve, is p, value we're trying to estimate.
01:06
I'm going to make an interval on this curve.
01:08
I'm going to put the middle of 95 % in this interval...