>When I tried this in Fox, I got .99 which is the correct standard deviation.
>So what is wrong with Excel? The STDEV function is actually an approximation of standard deviation which is designed for speed. For a "precise" version use STDEVP.
>
>If you have some free time here is the math:
>
>First calculate the average:
>=4.625
>
>Then for each value subtract the average and square it:
>0.390625 = (4-4.625)^2
>0.140625
>0.390625
>1.890625
>2.640625
>0.140625
>1.890625
>0.390625
>
>Finally total these values and square root them.
The Excel help gives different math, but I think the result should be the same. I understand, however, that the difference between the two versions is not "for speed", but to account for the fact that you are working only with a sample of a total population (and want to extrapolate the standard deviation for the entire population), vs., you have data for the entire population you are analyzing.
Hilmar.
Difference in opinions hath cost many millions of lives: for instance, whether flesh be bread, or bread be flesh; whether whistling be a vice or a virtue; whether it be better to kiss a post, or throw it into the fire... (from Gulliver's Travels)
