Anyone know what the 95% confidence
interval of the median would be?
For an R answer you could get one for each group from wilcox.test( ,
conf.int =TRUE ) and build that into an alternative boxplot stats function
which you could specify in your bwplot call.
Steve Ellison
Hi Bert, Dennis,
I'll agree that using a barchart was a poor choice. I was in fact using a
notched bwplot to show the median and confidence interval of the median. In
this case it's the median and confidence interval that I want to highlight,
and I find that the visual noise of the box and
On Jul 5, 2013, at 11:15 AM, Shaun Jackman wrote:
Hi Bert, Dennis,
I'll agree that using a barchart was a poor choice. I was in fact using a
notched bwplot to show the median and confidence interval of the median. In
this case it's the median and confidence interval that I want to
Be careful!
You are talking about 2 different varieties of apples here. As I read
it, the CI's in the cancer data, which I know is just for example
purposes, are CI's for the **individual means**; the notches in
boxplots are nonparametric and for 2 groups with roughly equal sample
sizes, The
Yes! Thank you, David. That's exactly what I'm I'm looking for. For
the record, here's a couple pages leading to this answer:
http://www.hep.by/gnu/r-patched/r-faq/R-FAQ_89.html
http://latticeextra.r-forge.r-project.org/man/segplot.html
Hmm. Interesting point, Bert. I don't know whether the notches show
the 95% confidence interval or the median, or the 95% confidence
interval that two non-overlapping notches have different medians.
You're saying it's the latter? Anyone know what the 95% confidence
interval of the median would be?
This is not an R question. Read the references.
Bert
Sent from my iPhone -- please excuse typos.
On Jul 5, 2013, at 12:15 PM, Shaun Jackman sjack...@gmail.com wrote:
Hmm. Interesting point, Bert. I don't know whether the notches show
the 95% confidence interval or the median, or the 95%
Shaun:
I understand that this type of plot is standard in many disciplines,
but it really is awful (google on 'Dynamite plots' for some more
erudite perspectives). Have you considered bwplot() for your
unaggregated data instead?
(No need to reply. It's July 4, and I'm just waving a little flag
8 matches
Mail list logo
