Should I use STDEV.S or STDEV.P?

[chasfh]

I have ten sets of data, each having ten data points within. I want to compare the standard deviation among all ten sets to see which has the widest distribution and which the narrowest.

My simple question is, should I use STDEV.S or STDEV.P for to calculate the standard deviation of each data set? I assume I should use P, since the entire population of each data set consists of the ten data points within it. But I wanted to check that off with the group here.

Please advise? Thanks.

 

[maabadi]

STDEV.S is Standard Deviation for Samples means small size
STDEV.P is Standard Deviation for Population means Large size
Because your sample size is small then You should use STDEV.S

 

[chasfh]

STDEV.S is Standard Deviation for Samples means small size
STDEV.P is Standard Deviation for Population means Large size
Because your sample size is small then You should use STDEV.S

Interesting. OK, thanks.

At what point do you think the size of the data set moves from "small sample" to "large population"?

 

[kweaver]

Your sample size can be small AND represent the entire population. The key issue is what does your data represent: a sample or the entire population?

 

[chasfh]

Your sample size can be small AND represent the entire population. The key issue is what does your data represent: a sample or the entire population?

My data sets are each ten data items. Ten is the entire population of each set.

Sounds like you may have a different idea about whether you use .S or .P as being dependent on the number of items, as maabadi suggests?

 

[kweaver]

Well, normally most people wanting S.D. tend to be interested in knowing the population standard deviation because the population contains all the values of interested. Therefore, you would normally calculate the population standard deviation if: (1) you have the entire population, large or small (in my view) or (2) you have a sample of a larger population, but you are only interested in this sample and do not wish to generalize your findings to the population. If your 10 are the entire population, then you said it: it's the population.