Anna Zakrisson
2013-Feb-19 11:38 UTC
[R] introducing jitter in overlapping graphs using ggplots (plotmeans). Also sciplot.
Hi,
I want to plot means with standard deviations of Total Nitrogen (TN) across
4 stations (S1-S4) and over 3 years (2007-2009). I want this plot in one
panel.
I do not want medians (bwplot, boxplot).
I have tried a few different packages and it seems that ggplots with
plotmeans was the fastest (I am not extremely skilled in writing my own
scripts). Unfortunately, there is no grouping argument in plotmeans().
Therefore, I have overlaid three plots (plotting all stations witin year).
However, as the graphs are exctly overlaid, it is not possible to
distinguish the confidence intervals. How do I introduce jitter?
Data2007, Data2008 and Data2009 are the dataframes with the data from that
particular year including all the stations.
I have also tried lineplot.CI from the package sciplot and also there I am
not able to introduce jitter. I have looked at the helpfiles.
PACKAGE GGPLOT:
library(gplots)
par(family="serif",font=1)
plotmeans(TN ~ STATION,
data =Data2007,
bars=TRUE, p=0.95,
pch=1,
cex=0.8,
n.label=FALSE,
mean.labels=FALSE,
barwidth=1.5, barcol="black",
connect=TRUE,
xlab="Station",
ylab = expression(paste("TN (",mu, g~L^{-1},
")")))
par(new=T)
plotmeans(TN ~ STATION,
data =Data2008,
bars=TRUE, p=0.95,
pch=2,
cex=0.8,
lty=2,
n.label=FALSE,
mean.labels=FALSE,
barwidth=1.5, barcol="black",
connect=TRUE,
na.action=na.exclude,
yaxt='n', ann=FALSE,
xaxt="n")
par(new=T)
plotmeans(TN ~ STATION,
data =Data2009,
bars=TRUE, p=0.95,
pch=3,
cex=0.8,
lty=3,
n.label=FALSE,
mean.labels=FALSE,
barwidth=1.5, barcol="black",
connect=TRUE,
na.action=na.exclude,
yaxt='n', ann=FALSE,
xaxt="n")
PACKAGE SCIPLOT:
library(sciplot)
lineplot.CI(response=TN, x.factor=STATION, group=YEAR,
ci.fun= function(x) c(mean(x)-sd(x), mean(x) + sd(x)),
data = Mydata,
xlab="Station",
ylab = expression(paste("TN(",mu, g~L^{-1},
")")))
#here I have calculated the standard deviations instead of the default
standard errors.
Thank you for taking your time. I have spent a long time trying to solve
this and the frustration slowly approaches anger (in myself) :-)
Yours sincerely
Anna Zakrisson Braeunlich
PhD Student
Department of Systems Ecology
Stockholm University
Svante Arrheniusv. 21A
SE-106 91 Stockholm
Lives in Berlin.
For fast paper-mail responses, please use the below address:
Katzbachstr. 21
D-10965, Berlin - Kreuzberg
Germany/Deutschland
E-mail: anna@ecology.su.se
Tel work: +49-(0)3091541281
Mobile: +49-(0)15777374888
Web site: http://www.ecology.su.se/staff/personal.asp?id=163
><((((º>`•. . • `•. .• `•. . ><((((º>`•. . • `•. .•
`•. .><((((º>
[[alternative HTML version deleted]]
John Kane
2013-Feb-19 15:07 UTC
[R] introducing jitter in overlapping graphs using ggplots (plotmeans). Also sciplot.
Hi Anna, A small point -- there is no package called ggplots. There is a package called gplots and one called ggplot2, which in earlier form was called ggplot. To see what is happening I believe we need some sample data from the three data files or some mock-up data that matches your actual data in form. The easiest way to supply data is to use the dput() function. Example with your file named "testfile": dput(testfile) Then copy the output and paste into your email. For large data sets, you can just supply a representative sample. Usually, dput(head(testfile, 100)) will be sufficient. In this case it looks like we would want three dput results, one for each year. John Kane Kingston ON Canada> -----Original Message----- > From: anna at ecology.su.se > Sent: Tue, 19 Feb 2013 12:38:25 +0100 > To: r-help at r-project.org > Subject: [R] introducing jitter in overlapping graphs using ggplots > (plotmeans). Also sciplot. > > Hi, > > I want to plot means with standard deviations of Total Nitrogen (TN) > across > 4 stations (S1-S4) and over 3 years (2007-2009). I want this plot in one > panel. > I do not want medians (bwplot, boxplot). > > I have tried a few different packages and it seems that ggplots with > plotmeans was the fastest (I am not extremely skilled in writing my own > scripts). Unfortunately, there is no grouping argument in plotmeans(). > Therefore, I have overlaid three plots (plotting all stations witin > year). > However, as the graphs are exctly overlaid, it is not possible to > distinguish the confidence intervals. How do I introduce jitter? > Data2007, Data2008 and Data2009 are the dataframes with the data from > that > particular year including all the stations. > > I have also tried lineplot.CI from the package sciplot and also there I > am > not able to introduce jitter. I have looked at the helpfiles. > > > PACKAGE GGPLOT: > > library(gplots) > par(family="serif",font=1) > plotmeans(TN ~ STATION, > data =Data2007, > bars=TRUE, p=0.95, > pch=1, > cex=0.8, > n.label=FALSE, > mean.labels=FALSE, > barwidth=1.5, barcol="black", > connect=TRUE, > xlab="Station", > ylab = expression(paste("TN (",mu, g~L^{-1}, ")"))) > par(new=T) > plotmeans(TN ~ STATION, > data =Data2008, > bars=TRUE, p=0.95, > pch=2, > cex=0.8, > lty=2, > n.label=FALSE, > mean.labels=FALSE, > barwidth=1.5, barcol="black", > connect=TRUE, > na.action=na.exclude, > yaxt='n', ann=FALSE, > xaxt="n") > par(new=T) > plotmeans(TN ~ STATION, > data =Data2009, > bars=TRUE, p=0.95, > pch=3, > cex=0.8, > lty=3, > n.label=FALSE, > mean.labels=FALSE, > barwidth=1.5, barcol="black", > connect=TRUE, > na.action=na.exclude, > yaxt='n', ann=FALSE, > xaxt="n") > > PACKAGE SCIPLOT: > library(sciplot) > lineplot.CI(response=TN, x.factor=STATION, group=YEAR, > ci.fun= function(x) c(mean(x)-sd(x), mean(x) + sd(x)), > data = Mydata, > xlab="Station", > ylab = expression(paste("TN(",mu, g~L^{-1}, ")"))) > > #here I have calculated the standard deviations instead of the default > standard errors. > > Thank you for taking your time. I have spent a long time trying to solve > this and the frustration slowly approaches anger (in myself) :-) > > Yours sincerely > > Anna Zakrisson Braeunlich > PhD Student > > Department of Systems Ecology > Stockholm University > Svante Arrheniusv. 21A > SE-106 91 Stockholm > > Lives in Berlin. > For fast paper-mail responses, please use the below address: > Katzbachstr. 21 > D-10965, Berlin - Kreuzberg > Germany/Deutschland > > E-mail: anna at ecology.su.se > Tel work: +49-(0)3091541281 > Mobile: +49-(0)15777374888 > Web site: http://www.ecology.su.se/staff/personal.asp?id=163 > > ><((((B:>`b?". . b?" `b?". .b?" `b?". . ><((((B:>`b?". . b?" `b?". .b?" > `b?". .><((((B:> > > [[alternative HTML version deleted]] > > ______________________________________________ > R-help at 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.____________________________________________________________ FREE ONLINE PHOTOSHARING - Share your photos online with your friends and family! Visit http://www.inbox.com/photosharing to find out more!
Jim Lemon
2013-Feb-19 21:06 UTC
[R] introducing jitter in overlapping graphs using ggplots (plotmeans). Also sciplot.
On 02/19/2013 10:38 PM, Anna Zakrisson wrote:> Hi, > > I want to plot means with standard deviations of Total Nitrogen (TN) across > 4 stations (S1-S4) and over 3 years (2007-2009). I want this plot in one > panel. > I do not want medians (bwplot, boxplot). >...Hi Anna, From your description, the brkdn.plot (plotrix) function might do what you want. In common with other functions in plotrix, brkdn.plot uses offsets rather than jittering to prevent overplotting. Jim
Reasonably Related Threads
- RSvgDevice & sapply(plotmeans)
- gplots
- Custom Plot - means, SD & 5th-95th% (Plotmeans or Boxplot)?
- How to plot 2 continous variables on double y-axis with 2 factors: ggplot2, gplot, lattice, sciplot?
- Adjusting Yaxis (ylim) limits on a plotMeans(DV, IV1, IV2, error.bars="se")