Difference between STDEV and STDEVP (pure)
Author Message
Difference between STDEV and STDEVP (pure)

Can someone please teell me the difference between stdev
and stdevp when it comes to worksheet function. We are
working with averages and would like to know which one to
use. Thank you,

Chris

Mon, 22 Aug 2005 08:20:28 GMT
Difference between STDEV and STDEVP (pure)

Stdev subtracts 1 from the number of observations, stdevp
doesn't.  The difference is whether you are calculating
the standard deviation of a data set the represents the
entire popluation (stdevp) or a sample of the population
(stdev)

Quote:
>-----Original Message-----
>Can someone please teell me the difference between stdev
>and stdevp when it comes to worksheet function. We are
>working with averages and would like to know which one to
>use. Thank you,

>Chris
>.

Mon, 22 Aug 2005 08:44:57 GMT
Difference between STDEV and STDEVP (pure)
Chris,
As  PC noted, the difference has to do with a divisor at a certain
point. When correctly calculated, they differ by a factor of
Sqrt((n-1)/n), where n is the sample size.

STDEV is a so-called "best linear unbiased estimator" of the "true"
standard deviation of an underlying population from which the sample was
drawn.  If you are working with data from the entire population, use
STDEVP.  If you are working with a randomly selected sample from a
larger population, use STDEV.

Well, that is common wisdom, at least.

That you write "working with averages", though, puzzles me.

BTW, is you aren't sure of which of these to use, ask yourself if the
difference in returned vales matters for the decision at hand.

HTH

Quote:

> Can someone please teell me the difference between stdev
> and stdevp when it comes to worksheet function. We are
> working with averages and would like to know which one to
> use. Thank you,

> Chris

--

Mon, 22 Aug 2005 09:14:28 GMT

 Page 1 of 1 [ 3 post ]

Relevant Pages