00:01
If we have a sample size of 100, a mean of 48, and a population standard deviation of 5, to get a confidence interval, i'm going to do mean plus or minus z alpha over 2 times standard deviation divided by square root n.
00:23
At 95 % confidence, my z alpha over 2 is z 0 .025, which equals 1 .96.
00:32
And at 99%, we have z 0 .005, which is 2 .58.
00:43
And both those values are found in the standard normal table.
00:49
So for my 95 % interval, i'll have 48 plus or minus 1 .96 times 5 over square root 100, which gives an interval of 47 .02 to 48 .98.
01:08
And for 99%, we'll have 48 plus or minus 2 .58 times 5 over square root 100, which gives an interval of 46 .71 to 49 .29.
01:29
So we should see as confidence goes up, the interval widens.
01:34
For part b, if i need to find my sample size, i'll use the formula n z alpha over 2 times standard deviation over error squared.
01:48
I'm doing a 95 % interval, so 1 .96 times our standard deviation 5.
01:56
And my desired error is 0 .5, which gives me n equals 384 .16 or about 385.
02:10
For part 2, if i now have a sample of 36 with a mean of 48 .5 and a standard deviation of 5, my null hypothesis is the mean equals 50...