s = (s1+s2+ ...sn)/n
Variance =[ (s1 - s)^2 + (s2 - s)^2
+ ... (sn - s)^2]/n
Standard Deviation = Variance^(1/2)
In excel: Variance is calculated by the function VAR(C1:Cn), and Standard Deviation
is calculated by STDEV(C1:Cn)
STDEV.S assumes that its arguments are a
sample of the population. If your data represents the entire population,
then compute the standard deviation using STDEV.P.
If
I remember correctly, STDEV.S in invented to get a more accurate result
for sampling from a large pool, when you don't have the whole
population or samples.
http://www.mathsisfun.com/data/standard-deviation.html
