[R] multiple x factors using the sciplot package

2010-08-25 Thread Jordan Ouellette-Plante

Dear R community,

 

I am a beginner using the sciplot package to graph barplots. I would like to be 
able to graph two x factors (Sampling.year and Period). 

Sampling.year: 2006, 2007, 2008, 2009

Period: First, Second, Total

 

The parameter group is the different species I looked at. They can be seen in 
the legend.

The parameter response is the Average percentage cover category of these 
species.

 

I would like the graph so you see on the x axis the years (in a bigger font) 
and the period (in smaller font). It would look a bit like the barplot that can 
be seen at 
http://onertipaday.blogspot.com/2007/05/make-many-barplot-into-one-plot.html, 
but with the advantage of the sciplot package.

 

 

Thanks a lot for your help!

 

Jordan
  
[[alternative HTML version deleted]]

__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.


Re: [R] multiple x factors using the sciplot package

2010-08-25 Thread Dennis Murphy
Hi:

You can probably do what you want in either ggplot2 or lattice, but I would
recommend at least a couple different approaches:

(1) Plot individual bar charts by combinations of year and period.
 This is easy to do in both ggplot2 and lattice: in ggplot2, one
 would use  geom_bar(x)  + facet_grid(year ~ period), where
 x is the count variable. The same thing in lattice would be
 barchart( ~ x | year * period, data = mydata). Suitable options
 can be added to either plot for better presentation. All the plots
 are on the same horizontal and vertical scales so that visual
 comparisons across the rows and columns are easy to make.
(2) Use a Cleveland dot chart. This is much less wasteful of ink
  than bar charts, and multiple groups can be compared without
  too much difficulty. See dotplot() in lattice for details. The
  VADeaths example is likely the best one to emulate for your
  problem. Both the input data structure and plot code matter.
(3) If your purpose is to compare the two groups with respect
  to a count variable, a third option is to consider a mosaic plot,
  which can be thought of as a 'visual chi-square test'. These
  can be found in package vcd, which contains a nice 40+
  page vignette to show you in detail how to create a suitable
  mosaic plot (and how to interpret it).

I prefer to see R used to promote good graphics and sound statistical
principles. In that spirit, performing a two-factor comparison with count
data is easier, and quite possibly more instructive, with a 2D visual
metaphor (which all three of the suggestions above have in common) than with
the 1D metaphor of multiple dodged/side-by-side bar charts. Of course, this
may be against the standard practice in your field, in which case you have a
decision to make.

HTH,
Dennis

On Wed, Aug 25, 2010 at 6:16 AM, Jordan Ouellette-Plante 
jordan_opla...@hotmail.com wrote:


 Dear R community,



 I am a beginner using the sciplot package to graph barplots. I would like
 to be able to graph two x factors (Sampling.year and Period).

 Sampling.year: 2006, 2007, 2008, 2009

 Period: First, Second, Total



 The parameter group is the different species I looked at. They can be
 seen in the legend.

 The parameter response is the Average percentage cover category of these
 species.



 I would like the graph so you see on the x axis the years (in a bigger
 font) and the period (in smaller font). It would look a bit like the barplot
 that can be seen at
 http://onertipaday.blogspot.com/2007/05/make-many-barplot-into-one-plot.html,
 but with the advantage of the sciplot package.





 Thanks a lot for your help!



 Jordan

[[alternative HTML version deleted]]

 __
 R-help@r-project.org mailing list
 https://stat.ethz.ch/mailman/listinfo/r-help
 PLEASE do read the posting guide
 http://www.R-project.org/posting-guide.html
 and provide commented, minimal, self-contained, reproducible code.


[[alternative HTML version deleted]]

__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.