Skip to content

Split

3 messages · Medic, Bert Gunter, Miloš Žarković

#
I have
mydata$var
#this is ONE group of patients

And I would like to get
median and ICR of mydata$var.

How can I get this?
With summary (mydata$var)!

Ok!

And now I would like to get THE SAME, but for TWO group: male and
female (which are contained in the group mydata$var)

How can I get this?
First I need to split mydata$var by mydata$sex, and then take:

summary (for male)
and
summary (for female)

That's all I want

Bert,
ave(mydata$var, madata$sex, FUN=median)
gives me:
[1] 5.6 5.6 5.6 5.6 5.6 5.6 5.6 5.6 5.6 5.6 5.6 5.6 5.6 5.6 5.6 5.6
5.6 5.6 5.6 5.6
[21] 6.0 6.0 6.0 6.0 6.0 6.0 6.0 6.0 6.0 6.0 6.0 6.0 6.0 6.0 6.0 6.0
6.0 6.0 6.0 6.0
What is it?
It is an endless(???) repetition of the median.
Moreover, there is no ICR.

Jeff,
your constructions are too complicated for me
===
P.S. Such simple thing and so difficult?! (I begin think about the Excel.)
#
Do you mean IQR? -- I don't know what ICR means.
If so, see IQR.

More generally see ?by or more generally ?tapply to obtain whatever sort of
summary you want.

e.g.
d$w: a
   median       IQR
0.5469662 0.4548506
--------------------------------------------------------------
d$w: b
   median       IQR
0.6860975 0.3456893


-- Bert
Bert Gunter

"The trouble with having an open mind is that people keep coming along and
sticking things into it."
-- Opus (aka Berkeley Breathed in his "Bloom County" comic strip )
On Mon, Dec 23, 2019 at 9:15 AM Medic <mailiPadpost at gmail.com> wrote:

            

  
  
#
J have just seen your follow-up post (out of tread). I don?t want to be
rude or patronizing but few caveats.

Neither R nor R help are meant to be user friendly. Learning curve is steep
but very rewarding at the end.

Problem you have can be solved in a literary hundred ways. Unfortunately,
your question was not a reproducible example and you got most complicated
answer. Apply entourage is kind of dark art in R, so I offered different
solutions, and hope that my answer helped.

Being medic too, I understand your pain, but for the star I suggest Rcmdr.


Regards,


Milo?
On Mon, 23 Dec 2019 at 20:36, Bert Gunter <bgunter.4567 at gmail.com> wrote: