Creating graphs

“No design works unless it embodies ideas that are held common by the people for whom the object is intended.” - Adrian Forty

The use of R offers many advantages for data analysis. R is also a useful tool for creating scientific graphs. R makes it possible

• to produce high quality graphs that are also suitable for professional printing.
• to create graphics completely in R syntax and thus make them reproducible.
• to flexibly combine different types of graphics or develop your own forms of presentation.
• to easily create uniformly designed graphics.

Do not wait until the very final analysis stage to produce some publication quality graphics but produce fast (not necessarily nice) visualizations all the way through your data analysis. Otherwise you will not utilize the best neural network infrastructure you have available: your brain and its ability to identify patterns.

Basic plots in R

plot(x,y) is the universal function for generating scatter plots and line plots from the vectors x and y. plot can be optionally adjusted with various parameters:

Parameter	Description	Example
type	sets the diagram type: “l” -> lines “p” -> dots “b” -> both points and lines “c”-> empty points joined by lines “h” -> bars, histogram-like vertical lines “s” -> step diagram “n” -> no representation, does not produce any points or lines	type=”b”
main	name of title	main=”Population growth (annual %)”
col	set color for lines, dots, bars etc. show available color words: colours()	col=”blue” col=”#112233” (RGB value) col=”#11223344” (RGB and transparency value)
sub	subtitle, caption	sub=”Figure 1”
xlab	labeling of the x-axis	xlab=”Years 2000 to 2020”
ylab	labeling of the y-axis	ylab=”Population Growth in %”
pch	The graphic symbol used to represent data points. Either a character as a string or a number representing the following characters;	pch=1
lty	Stroke type of lines	lty=1
lwd	Line width	lwd=3
cex	Symbol size	cex=2
xlim	Sets start and end values of the X-axis scale	xlim=c(0,20)
ylim	Sets start and end values of the y-axis scale	ylim=c(100,160)
bty	Outline of the complete character area. Options are: “n”, “o”, “c”, “l”, “7”	bty=”n”
xaxt	Labeling of the X-axis scale OFF	xaxt=”n”
yaxt	Labeling of the Y-axis scale OFF	yaxt=”n”
font	Selection of the font format, 1=plain, 2=bold, 3=italic, 5=Greek symbols	font=5
	“Pointcharacters as used in GNU R” by Marc Schwenzer on Wikimedia “Line-Styles of GNU R” by Europol on Wikimedia

Text key for the symbol charts: pch = 0 is an open square, 1 an open circle, 2 an open triangle, 3 a plus, 4 a cross, 5 an open diamond, 15 a filled square, 16 a filled circle and 17 a filled triangle. Line types lty = 1:6 are solid, dashed, dotted, dot-dash, long-dash and two-dash, respectively.

Let’s have some fun

The following is an introduction for producing simple graphs.

Define 2 vectors for two different datasets:

sport_cars <- c(1, 2, 4, 2, 6, 7, 10)
family_cars <- c(3, 6, 6, 8, 6, 5, 2)

Then calculate the range from 0 to max value of sport cars and family cars:

max_range <- range(0, sport_cars, family_cars)

We graph the cars using the y axis that ranges from 0 to max value in sports or family cars and turning off axes and annotations (axis labels) so we can specify them ourselves.

plot(sport_cars, type="o", col="blue", ylim=max_range,
     axes=FALSE, ann=FALSE)

Print of Code chunk above

Make x axis using Mon-Fri labels

axis(1, at=1:7, lab=c("Mon","Tue","Wed","Thu","Fri","Sat","Sun"))

Print of Code chunk above

Make y axis with horizontal labels that display ticks at every 4 marks. 4*0:g_range[2] is equivalent to c(0,4,8,12).

axis(2, las=1, at=4*0:max_range[2])

Print of Code chunk above

Create box around plot

box()

Print of Code chunk above

Graph trucks with red dashed line and square points

lines(family_cars, type="o", pch=22, lty=2, col="red")

Print of Code chunk above

Create a title with a red, bold/italic font

title(main="Car Types on Streets", col.main="red", font.main=4)

Print of Code chunk above

Label the x and y axes with dark green text

title(xlab="Days", col.lab=rgb(0,0.5,0))
title(ylab="Total", col.lab=rgb(0,0.5,0))

Print of Code chunk above

Create a legend at (1, g_range[2]) that is slightly smaller (cex) and uses the same line colors and points used by the actual plots

legend(1, max_range[2], c("Sport","Family"), cex=0.8,
       col=c("blue","red"), pch=21:22, lty=1:2);

Print of Code chunk above

In total this is the code and the output in the plot-window will look like this with each line:

sport_cars <- c(1, 2, 4, 2, 6, 7, 10)
family_cars <- c(3, 6, 6, 8, 6, 5, 2)
max_range <- range(0, sport_cars, family_cars)
plot(sport_cars, type="o", col="blue", ylim=max_range,
     axes=FALSE, ann=FALSE)
axis(1, at=1:7, lab=c("Mon","Tue","Wed","Thu","Fri","Sat","Sun"))
axis(2, las=1, at=4*0:max_range[2])
box()
lines(family_cars, type="o", pch=22, lty=2, col="red")
title(main="Car Types on Streets", col.main="red", font.main=4)
title(xlab="Days", col.lab=rgb(0,0.5,0))
title(ylab="Total", col.lab=rgb(0,0.5,0))
legend(1, max_range[2], c("Sport","Family"), cex=0.8,
       col=c("blue","red"), pch=21:22, lty=1:2);

Animation of the plot process

Visualizations 101

No matter if you produce publication quality graphics or quick and dirty visualizations, remember some guiding principles and traps illustrated in the upcoming figures.

For further reading, have a look at e.g. Kelleher & Wagner 2011. To get some inspiration, the R Graph Gallery is also a good place to start (but it relies on the ggplot2 library, not on the basic R plotting functions).

Test your knowledge!