[Rd] Documentation for sd (stats) + suggestion

PatrickT @nnoporc| @end|ng |rom gm@||@com
Tue Feb 19 10:16:48 CET 2019


I cannot file suggestions on bugzilla, so writing here.

As far as I can tell, the manual help page for ``sd``

?sd

does not explicitly mention that the formula for the standard deviation is
the so-called "Bessel-corrected" formula (divide by n-1 rather than n).

I suggest it should be stated near the top.

I would also suggest (feature request!) that either

 - a population standard deviation formula, e.g. ``sdp`` or ``sd.p`` be
made available (that would be my preference)

or

 - the current ``sd`` be extended to accept a ``population=FALSE`` or
``sample=TRUE`` argument.

Same for the variance. Excel, Calc, etc. offer these.

Motivation: I encourage my students to use R rather than Python (which has
picked up big time in recent years) on the grounds that it is easier to get
started with and is specialized in statistics. But then there is no
"population" formula for the standard deviation. And ``mode`` is not the
mode they expect... (btw I suggest adding a ``modes`` function to the
core)  All things a beginner will look for in a stats software.

Thanks for listening. And thanks for the great work!

	[[alternative HTML version deleted]]



More information about the R-devel mailing list