DescTools (version 0.99.18)

PlotDot: Cleveland's Dot Plots

Description

Draw a Cleveland dot plot. This is an extended version of dotchart with an added option for error bars and an add argument.

Usage

PlotDot(x, labels = NULL, groups = NULL, gdata = NULL, cex = par("cex"), pch = 21, gpch = 21, bg = par("bg"), color = par("fg"), gcolor = par("fg"), lcolor = "gray", xlim = NULL, main = NULL, xlab = NULL, ylab = NULL, add = FALSE, args.errbars = NULL, ...)

Arguments

x
either a vector or matrix of numeric values (NAs are allowed). If x is a matrix the overall plot consists of juxtaposed dotplots for each row. Inputs which satisfy is.numeric(x) but not is.vector(x) || is.matrix(x) are coerced by as.numeric, with a warning.
labels
a vector of labels for each point. For vectors the default is to use names(x) and for matrices the row labels dimnames(x)[[1]].
groups
an optional factor indicating how the elements of x are grouped. If x is a matrix, groups will default to the columns of x.
gdata
data values for the groups. This is typically a summary such as the median or mean of each group.
cex
the character size to be used. Setting cex to a value smaller than one can be a useful way of avoiding label overlap. Unlike many other graphics functions, this sets the actual size, not a multiple of par("cex").
pch
the plotting character or symbol to be used.
gpch
the plotting character or symbol to be used for group values.
bg
the background color of plotting characters or symbols to be used; use par(bg= *) to set the background color of the whole plot.
color
the color(s) to be used for points and labels.
gcolor
the single color to be used for group labels and values.
lcolor
the color(s) to be used for the horizontal lines.
xlim
horizontal range for the plot, see plot.window, e.g.
main
overall title for the plot, see title.
xlab, ylab
axis annotations as in title.
add
logical specifying if bars should be added to an already existing plot; defaults to FALSE.
args.errbars
optional arguments for adding error bars. All arguments for ErrBars can be supplied. If left to NULL (default), no error bars will be plotted.
...
graphical parameters can also be specified as arguments.

Value

Details

This function is invoked for its side effect, which is to produce two variants of dotplots as described in Cleveland (1985).

Dot plots are a reasonable substitute for bar plots.

References

Becker, R. A., Chambers, J. M. and Wilks, A. R. (1988) The New S Language. Wadsworth & Brooks/Cole.

Cleveland, W. S. (1985) The Elements of Graphing Data. Monterey, CA: Wadsworth.

Murrell, P. (2005) R Graphics. Chapman & Hall/CRC Press.

Examples

Run this code
PlotDot(VADeaths, main = "Death Rates in Virginia - 1940")
op <- par(xaxs = "i")  # 0 -- 100%
PlotDot(t(VADeaths), xlim = c(0,100),
         main = "Death Rates in Virginia - 1940")
par(op)

# add some error bars
PlotDot(VADeaths, main="Death Rates in Virginia - 1940", col="red", pch=NA,
        args.errbars = list(from=VADeaths-2, to=VADeaths+2, mid=VADeaths,
        pch=21, cex=1.4))

# add some other values
PlotDot(VADeaths+3, pch=15, col="blue", add=TRUE, labels=NA)

# same as PlotDotCI
xci <- do.call(rbind, tapply( d.pizza$delivery_min, d.pizza$driver,
               MeanCI, conf.level=0.99, na.rm=TRUE))

PlotDot(xci[,1], main="delivery_min ~ driver",
        args.errbars = list(from=xci[,2], to=xci[,3], mid=xci[,1],
        lwd=2, pch=21, col="grey40", bg.pch="grey50", col.pch="black"),
        xlim=c(15,35), panel.before=grid())

# with group data
x <- with(d.pizza, tapply(temperature, list(area, driver), mean, na.rm=TRUE))

PlotDot(x, gdata = tapply(d.pizza$temperature, d.pizza$driver, mean, na.rm=TRUE),
        gpch = 15)

# special format
par(lend=1)

PlotDot(VADeaths, main="Death Rates in Virginia - 1940", pch=NA, lcolor = hecru,
        args.errbars = list(from=VADeaths-2, to=VADeaths+2, mid=VADeaths,
                            pch="|", col.pch=hred, cex=1.3, lwd=8, code=0, col=hgreen))

# Error bars for binomial confidence intervals
tab <- table(d.pizza$driver, d.pizza$wine_delivered)
xci <- SetRowNames(BinomCI(tab[,1], rowSums(tab)), rownames(tab))
PlotDot(xci[,1], main="wine delivered ~ driver ", xlim=c(0,1),
        args.errbars=list(from=xci[,-1], mid=xci[,1], pch=21))


# Error bars for confidence intervals for means
xci <- do.call(rbind, tapply(d.pizza$delivery_min, d.pizza$driver,
  MeanCI, conf.level=0.99, na.rm=TRUE))

PlotDot(xci[,1], main="delivery_min ~ driver",
  args.errbars=list(from=xci[,-1]))

# Setting the colours
# define some error bars first
lci <- sweep(x = VADeaths, MARGIN = 2, FUN = "-", 1:4)
uci <- sweep(x = VADeaths, MARGIN = 1, FUN = "+", 1:5)

PlotDot(VADeaths, main="This should only show how to set the colours, not be pretty",
        col=c("red", "blue"), pch=NA,
        gcolor = c("green", "blue", "orange", "magenta"), gdata=c(10,20,30,40),
        gpch = c(15:18), lcolor = "orange",
        args.errbars = list(from=lci, to=uci, mid=VADeaths,
                            pch=21, cex=1.4, col="blue", col.pch="green", bg.pch="red"))

Run the code above in your browser using DataCamp Workspace