Hi all, is there any way in statistic node, or maybe another node to configure if you want to calculate the standard deviation in a sample or in a population? (basically apply or not Bessels correction)

In excel and google sheets for example, you can choose *stdev* or *stdevp*

Thanks is advance for your help

Hello @guillermoalvma,

And welcome to our community. Thank you for sharing your first contribution.

Before we probably get into more detail I want to ask, if you already used the *<math formula>* node, which gives you a variety of mathematical functions, as well as COL_STDEV. This basically refers to *stdev.p*. You could compute the *stdev.s* by creating a sample of your data set (e.g. with *<row sampling>* or *<partitioning>* node).

Would that be a start for a discussion?

Best regards,

Kevin

Thanks a lot for your answer and for the welcome.

The COL_STDEV function in math formula node calculates *stdev.s*, i.e the node assume that the dataset is a sample, not a population, that’s the problem.

Best regards.

Hi, is there any other info that you need to continue the discussion?

Hello @guillermoalvma,

Sorry for coming back to you topic just now.

As it turns out, there was a misunderstanding about the function inside the *<math formula>* node and you were right about the fact, that it indeed calculates *stdev.s*.

Since there is no additional node or function available to compute *stdev.p* right away, one needs to recreate its formula with a series of *<math formula>* nodes. But this is just one way to do it.

I will open a development ticket on our side to allow calculating it with a simple function for a future release though.

Best regards,

Kevin

This topic was automatically closed 7 days after the last reply. New replies are no longer allowed.