00:01
We're looking at a survey on the death penalty.
00:03
Half of the people were asked, do you believe it is applied fairly or unfairly? the other half were asked, is it unfairly or fairly? what kind of bias we're seeing here is we get different results? this is a form of response bias, and specifically it is order bias.
00:23
People respond differently depending on what order the questions or the outcomes come in.
00:30
In this case, the possible options they could choose came in different orders, and that affected the responses.
00:39
Part b, we're now going to combine them into an overall group.
00:43
So now our overall group has a sample size of 1 ,020, and the sample proportion, who said it was fairly applied, so 58 % for half of them, 54 % for half of them, the average is going to be between them.
01:02
We don't need to take a weighted average in this case.
01:06
If the sample sizes for each group were different, you would have to do a weighted average.
01:11
But since they're the same here, i'm just going halfway between.
01:14
So this is the proportion of this overall group who thought it was fairly applied.
01:19
We need to take a 95 % confidence interval.
01:22
So the formula for a confidence interval, for a proportion, when you have a nice big sample size, is p -hat, your point estimate, plus and minus your margin of error, which is z root p hat 1 minus p hat over n.
01:40
This is the formula we're going to use.
01:42
We have n, we have p hat, we get z from the level of confidence.
01:47
So z is from the normal distribution.
01:51
And when you have a proportion with a nice big sample size like this, you have an approximately normal approximation to it.
02:01
So here's my normal approximation.
02:03
I have my p hat in the middle.
02:04
And i'm saying, i'm going to put down a confidence interval where 95 % of sample proportions and the original population proportion is in here somewhere, 95 % of values of this curve are in my interval.
02:18
5 % are in the tails.
02:20
This is symmetric.
02:22
So each tail is 2 .5%...