00:01
We want to know the proportion of a workforce with two or more jobs.
00:06
But we don't know what sample size is required for the accuracy we want.
00:10
Imagine if we did know the sample size.
00:13
Well, i could take the sample, find p -hat, the sample proportion, and make a confidence interval to estimate the population proportion.
00:21
My formula would be p -hat, point estimate, plus and minus the margin of error, z root p -hat, 1 minus p -hat, over n.
00:31
I'm just going to focus on this margin of error, since we're told we should get the estimate to be within 1%.
00:37
I'm putting that in decimal form because we're going to do some calculations and percentages do not play well with calculations.
00:45
Now let's rearrange and solve for n.
00:47
Then we'll know the sample size necessary to make this true.
00:49
So we'll divide by z, square both sides, multiply by n, divide by this.
00:59
There we go.
01:00
We just need p -hat and z.
01:02
So for p -hat, we have been given a previous estimate.
01:06
A pilot survey found that 8 out of 40 met the criteria, a proportion of 0 .2.
01:14
So we're going to use that as our planning value.
01:17
We get z from the level of confidence.
01:20
Initially, this is going to be a binomial experiment.
01:23
N independent trials, two outcomes, each person either has two or more jobs or they don't...