Re: The calculator is correct Message #9 Posted by rsenzer on 25 July 2002, 7:56 p.m., in response to message #8 by Xtian
Hi Xtian,
I can't answer directly because I don't have a 42S. However, I have performed the calculations on a 32SII. The result of squaring the sample standard deviation on this calculator is :
174.999999999
So I think it's just a matter of rounding error.
The more important issue is distinguishing between sample statistics, which use a denominator of (N - 1) based on loss of a degree of freedom, and the population statistics, which use a denominator of N. [N being the number of data points.] On the 32SII, the sample standard deviation is given by sx whereas the population standard deviation is given by
[lower case greek letter] sigmax.
If a calculator has only one or the other, then
sigmax2 =
[(N - 1) / N] * sx2
and
sx2 =
[N / (N - 1)] * sigmax2
Just take the square root to compute sigmax or sx.
|