xBar ± z * sx / sqrt(n)
where xBar is the sample
mean
z is the zscore for having
α% of the data in the tails, i.e., P( |Z| > z) = α
sx is the sample standard
deviation
n is the sample size
For sample size
calcualtions we need the following. The width of the interval, from one end
point to the center is:
z * sx / sqrt(n) =
w and solving for n gives:
<span> n = (z * sx / w) ^ 2 </span>
remember that n needs to
be an integer. Always take the ceiling, i.e., round up. If you round down then
the width of the interval will not be correct, it will be too wide. By rounding
up, the interval will be more narrow than asked for, but this is a good thing.
It means there is more precision in the estimate.
<span> here we have </span>
z = 2.05
w = 0.04 * 152 = 6.08
n = (z * sx / w) ^ 2 n =
(2.05 * 26 / 6.08)^2
n = 76.8506
Round of the answer since
n must be an integer.
<span>n = 77 </span>