RFC: Statistics in Subsurface

[Dirk Hohndel]

The video that Pedro linked to seemed to indicate that the first chart is most likely to be understood, and that the the second one was harder to see trends in.

Conflicting with that is the desire to more typically show bar graphs sideways, as that makes it easier to deal with many data sets (think labeling the columns vs. labeling rows)

So all this is super helpful in figuring out how we should visualize these things - but not necessarily all leading to the same answers, as I'm not sure how well these line graphs work when turned 90 degrees :-)

/D



> On May 13, 2020, at 12:53 PM, Hartley Horwitz <hhrwtz at gmail.com> wrote:
> 
> I"ve attached 3 graphs showing the statistics summary.  Once again I showed them to a work colleague.  He found the upper 2 graphs easiest to understand.  
> 
> ...Hartley
> 
> On Wed, May 13, 2020 at 3:24 PM Dirk Hohndel <dirk at hohndel.org <mailto:dirk at hohndel.org>> wrote:
> That is excellent input!
> 
> Your final point is one that I had kinda assumed - most of the "more interesting" data no one but a geek will look into. And to them either box and whiskers (so quartiles) or at least floating box with mean (or your version in the first SAC chart below with the 0 based box with the mean as height and with whiskers for min/max) should make sense. But it also makes sense to look for simper ways to give access to the same data. Can you give an example for the "line graph with 3 lines for min/mean/max"?
> 
> Thanks
> 
> /D

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.subsurface-divelog.org/pipermail/subsurface/attachments/20200513/b379dc81/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: simple_stats.JPG
Type: image/jpeg
Size: 37710 bytes
Desc: not available
URL: <http://lists.subsurface-divelog.org/pipermail/subsurface/attachments/20200513/b379dc81/attachment-0001.jpe>