Hi!
I have 2 problems in drawing a stacked bar plot:
(1) This is a stacked bar plot with more than 100 bars next to each
other. So there should not be names at the bottom of the bars because
the bars are too narrow. I tried arg.names=NULL but that does not
work because R uses the row names from the data.frame.
How can I suppress the placing of names below the bars?
(2) The position of the legend is problematic. The legend is big, more
than 20 items. It overlays the bars themselves and makes the plot
unreadable. I would like to move it outside the plot and have tried
legend("topleft") but R complains about missing information.
Here is what I have so far
########################################################
barplot(t(file.codes),
+ beside = FALSE,
+ col = rainbow(ncol(file.codes)), names.arg=NULL,
+ legend = FALSE,
+ main = "Presidential documents - Codes per document",
+ xlab = "document", ylab = "number of codings")
#######################################################
Any ideas?
Thanks,
Henri-Paul
--
Curriculum & Instruction
Texas A&M University
TutorFind Learning Centre
Email: hindiogine at gmail.com
Skype: hindiogine
Website: http://people.cehd.tamu.edu/~sindiogine
On 20.10.2011 18:03, Henri-Paul Indiogine wrote:> Hi! > > I have 2 problems in drawing a stacked bar plot: > > (1) This is a stacked bar plot with more than 100 bars next to each > other. So there should not be names at the bottom of the bars because > the bars are too narrow. I tried arg.names=NULL but that does not > work because R uses the row names from the data.frame. > > How can I suppress the placing of names below the bars?It is the argument *names.arg* and it has to be set to NA.> (2) The position of the legend is problematic. The legend is big, more > than 20 items. It overlays the bars themselves and makes the plot > unreadable. I would like to move it outside the plot and have tried > legend("topleft") but R complains about missing information.arrange it outside by, e.g. increasing the size of margins (see argument "mar" in ?par) and place a separate legend (see ?legend) into the margins (see xps argument in ?par).> Here is what I have so far > > ######################################################## > barplot(t(file.codes), > + beside = FALSE, > + col = rainbow(ncol(file.codes)), names.arg=NULL, > + legend = FALSE, > + main = "Presidential documents - Codes per document", > + xlab = "document", ylab = "number of codings") > #######################################################Not reproducible since we do not have file.codes. Uwe Ligges> Any ideas? > > Thanks, > Henri-Paul > >
Hi Uwe! Thanks for the feedback. However, now the graph does not draw the y-axis nor labels and gives the following error message: Error in barplot.default(t(file.codes), beside = FALSE, names.arg = NA, : incorrect number of names This is the R code now: barplot(t(file.codes), beside = FALSE, names.arg=NA, legend = FALSE, main="test stacked bar plot", xlab="documents", ylab="number of codes", col=rainbow(ncol(file.codes))) I have head(file.codes) at the bottom of this email. Thanks, HP 2011/10/20 Uwe Ligges <ligges at statistik.tu-dortmund.de>:> It is the argument *names.arg* and it has to be set to NA.> head(file.codes)Achievement Diversity Economy EducEquity Gap HumanBenefit DCPD-200900575.scrb 2 2 4 5 2 1 DCPD-200900595.scrb 2 0 12 0 2 1 DCPD-200900884.scrb 1 0 12 0 1 1 DCPD-201000036.scrb 2 0 1 0 1 0 DCPD-201000130.scrb 1 0 1 0 2 0 DCPD-201000636.scrb 4 0 10 3 2 1 InternatComp MathSciEng Poverty Standards TechnoSociety DCPD-200900575.scrb 1 1 2 2 4 DCPD-200900595.scrb 1 2 0 7 7 DCPD-200900884.scrb 4 2 0 9 7 DCPD-201000036.scrb 5 1 0 2 2 DCPD-201000130.scrb 1 1 0 1 2 DCPD-201000636.scrb 2 2 0 8 1 WorkCareer DropOut Accountability AssessStudent Funding DCPD-200900575.scrb 7 0 0 0 0 DCPD-200900595.scrb 9 2 0 0 0 DCPD-200900884.scrb 12 0 1 2 0 DCPD-201000036.scrb 5 0 0 0 1 DCPD-201000130.scrb 2 0 0 0 1 DCPD-201000636.scrb 12 0 2 1 0 Globalization AllStudents NationInterest ProfDev DCPD-200900575.scrb 0 0 0 0 DCPD-200900595.scrb 0 0 0 0 DCPD-200900884.scrb 0 0 0 0 DCPD-201000036.scrb 0 0 0 0 DCPD-201000130.scrb 0 0 0 0 DCPD-201000636.scrb 0 0 0 0 NationDefense BestFirst AdmissionCollege DCPD-200900575.scrb 0 0 0 DCPD-200900595.scrb 0 0 0 DCPD-200900884.scrb 0 0 0 DCPD-201000036.scrb 0 0 0 DCPD-201000130.scrb 0 0 0 DCPD-201000636.scrb 0 0 0 -- Henri-Paul Indiogine Curriculum & Instruction Texas A&M University TutorFind Learning Centre Email: hindiogine at gmail.com Skype: hindiogine Website: http://people.cehd.tamu.edu/~sindiogine
On 20.10.2011 18:58, Henri-Paul Indiogine wrote:> Hi Uwe! Thanks for the feedback. However, now the graph does not > draw the y-axis nor labels and gives the following error message: > > Error in barplot.default(t(file.codes), beside = FALSE, names.arg = NA, : > incorrect number of namesnames.arg = rep(NA, nrow(file.codes)) in that case. Best, Uwe Ligges> This is the R code now: > > barplot(t(file.codes), beside = FALSE, names.arg=NA, legend = FALSE, > main="test stacked bar plot", xlab="documents", ylab="number of > codes", col=rainbow(ncol(file.codes))) > > > I have head(file.codes) at the bottom of this email. > > Thanks, > HP > > 2011/10/20 Uwe Ligges<ligges at statistik.tu-dortmund.de>: >> It is the argument *names.arg* and it has to be set to NA. > >> head(file.codes) > > Achievement Diversity Economy EducEquity Gap HumanBenefit > DCPD-200900575.scrb 2 2 4 5 2 1 > DCPD-200900595.scrb 2 0 12 0 2 1 > DCPD-200900884.scrb 1 0 12 0 1 1 > DCPD-201000036.scrb 2 0 1 0 1 0 > DCPD-201000130.scrb 1 0 1 0 2 0 > DCPD-201000636.scrb 4 0 10 3 2 1 > InternatComp MathSciEng Poverty Standards TechnoSociety > DCPD-200900575.scrb 1 1 2 2 4 > DCPD-200900595.scrb 1 2 0 7 7 > DCPD-200900884.scrb 4 2 0 9 7 > DCPD-201000036.scrb 5 1 0 2 2 > DCPD-201000130.scrb 1 1 0 1 2 > DCPD-201000636.scrb 2 2 0 8 1 > WorkCareer DropOut Accountability AssessStudent Funding > DCPD-200900575.scrb 7 0 0 0 0 > DCPD-200900595.scrb 9 2 0 0 0 > DCPD-200900884.scrb 12 0 1 2 0 > DCPD-201000036.scrb 5 0 0 0 1 > DCPD-201000130.scrb 2 0 0 0 1 > DCPD-201000636.scrb 12 0 2 1 0 > Globalization AllStudents NationInterest ProfDev > DCPD-200900575.scrb 0 0 0 0 > DCPD-200900595.scrb 0 0 0 0 > DCPD-200900884.scrb 0 0 0 0 > DCPD-201000036.scrb 0 0 0 0 > DCPD-201000130.scrb 0 0 0 0 > DCPD-201000636.scrb 0 0 0 0 > NationDefense BestFirst AdmissionCollege > DCPD-200900575.scrb 0 0 0 > DCPD-200900595.scrb 0 0 0 > DCPD-200900884.scrb 0 0 0 > DCPD-201000036.scrb 0 0 0 > DCPD-201000130.scrb 0 0 0 > DCPD-201000636.scrb 0 0 0 > >
Hi Uwe! 2011/10/20 Uwe Ligges <ligges at statistik.tu-dortmund.de>:> ?names.arg = rep(NA, nrow(file.codes))Yes, that works beautifully. Danke schoen! Henri-Paul -- Henri-Paul Indiogine Curriculum & Instruction Texas A&M University TutorFind Learning Centre Email: hindiogine at gmail.com Skype: hindiogine Website: http://people.cehd.tamu.edu/~sindiogine