Email address obfuscation in effect -- please
click here to turn it off.
[
Date Prev][
Date Next][
Thread Prev][
Thread Next][
Date Index][
Thread Index]
On Fri, 8 Jun 2007, Jonathan King wrote:
On 6/8/07, Mike Miller <EMAIL:PROTECTED> wrote:
On Fri, 8 Jun 2007, Mike Miller wrote:
Here's one that uses color for groups, with an appropriate legend, error
bars, grid lines and reasonable x-axis labels.
http://addictedtor.free.fr/graphiques/graphiques/graph_54.png
That's decent, but note that the bar colors/patterns could have been
more carefully chosen. This is a two-by-two classification (sex and
urban/rural), but you can't get that from the bars they use. A common
way to do this would be (say) solid for urban and striped for rural, and
then two different colors for gender.
A very good suggestion, or it could be lighter/darker for rural/urban and
red/blue for female/male (to use colors without patterns).
It is surprisingly tricky to find the "best" way to plot data,
especially for people who are unfamiliar with your work.
That was one of the best bar graphs I could come up with from a google
image search. It still wasn't all that great. I like statistical graphs,
a lot, and I wish more people cared to do a good job with them. One graph
can convey a lot of information if done correctly.
Mike
_______________________________________________
discussion mailing list
EMAIL:PROTECTED
http://mlug.missouri.edu/mailman/listinfo/discussion
Home | FAQ | Server | Presentations | Mailing Lists/Archives | Member Tools | Links | Sponsors | Contact