GRAPHS AND CHARTS IN R

M. A. Iquebal
I.A.S.R.I., Library Avenue, New Delhi – 110012
[email protected]

R includes two different packages for plotting data: graphics and lattice. The graphics
package contains a wide variety of functions for plotting data. It is very easy to customize
or modify charts with graphics package, or to interact with plots on the screen. The lattice
package contains an alternative set of functions for plotting data. Lattice graphics are well
suited for splitting data by a conditioning variable. This chapter gives an overview of the
graphics package.

AN OVERVIEW OF R GRAPHICS

Diagrammatic representation can make strong impact on a reader because of its visual
effect. However, these plots and charts should be constructed carefully because visual
effect can be easily misread or misused. Some of the types of diagrams that are commonly
used are described below. R includes tools for drawing most common type of charts,
including bar charts, pie charts, line charts and scatter plots. Additionally, R can also draw
some less familiar charts like quantile-quantile (Q-Q) plots, mosaic plots and contour
plots. Each graphical function has a large number of options making the production of
graphics very flexible.

1. PLOT
Description
Generic function for plotting of R objects. For simple scatter plots, plot.default will be
used. However, there are plot methods for many R objects, including functions,
data.frames, density objects, etc. Use methods(plot) and the documentation for these.

Usage
plot(x, y, ...)

Arguments
x the coordinates of points in the plot. Alternatively, a single plotting structure,
function or any R object with a plot method can be provided.
y the y coordinates of points in the plot, optional if x is an appropriate structure.
type
what type of plot should be drawn. Possible types are
 "p" for points,
 "l" for lines,
 "b" for both,
 "c" for the lines part alone of "b",
 "o" for both ‘overplotted’,
 "h" for ‘histogram’ like (or ‘high-density’) vertical lines,
 "s" for stair steps,
 "S" for other steps, see ‘Details’ below,
 Graphs and Charts in R

 "n" for no plotting.

main :an overall title for the plot.

sub :a sub title for the plot.
xlab:a title for the x axis.
ylab:a title for the y axis.

2. SCATTER PLOTS

plot.default {graphics}
The default scatterplot function

Description
Draw a scatter plot with decorations such as axes and titles in the active graphics window.

Usage
plot(x, y = NULL, type = "p", xlim = NULL, ylim = NULL,log = "", main = NULL, sub
= NULL, xlab = NULL, ylab = NULL,ann = par("ann"), axes = TRUE, frame.plot =
axes,panel.first = NULL, panel.last = NULL, asp = NA, ...)

Arguments

x, y thex and y arguments provide the x and y coordinates for the plot. Any
reasonable way of defining the coordinates is acceptable. See the function
xy.coords for details. If supplied separately, they must be of the same length.
type 1-character string giving the type of plot desired. The following values are
possible, for details, "p" for points, "l" for lines, "o" for overplotted points and
lines, "b", "c") for (empty if "c") points joined by lines, "s" and "S" for stair
steps and "h" for histogram-like vertical lines. Finally, "n" does not produce
any points or lines.
xlim the x limits (x1, x2) of the plot. Note that x1 > x2 is allowed and leads to a
‘reversed axis’.
The default value, NULL, indicates that the range of the finite values to be
plotted should be used.
ylim the y limits of the plot.
log a character string which contains "x" if the x axis is to be logarithmic, "y" if the
y axis is to be logarithmic and "xy" or "yx" if both axes are to be logarithmic.
main a main title for the plot, see also title.
sub a sub title for the plot.
xlab a label for the x axis, defaults to a description of x.
ylab a label for the y axis, defaults to a description of y.
ann a logical value indicating whether the default annotation (title and x and y axis
labels) should appear on the plot.
axes a logical value indicating whether both axes should be drawn on the plot. Use
 Graphs and Charts in R

graphical parameter"xaxt" or "yaxt" to suppress just one of the axes.

frame.plot a logical indicating whether a box should be drawn around the plot.
panel.first an ‘expression’ to be evaluated after the plot axes are set up but before any
plotting takes place. This can be useful for drawing background grids or
scatterplot smooths. Note that this works by lazy evaluation: passing this
argument from other plot methods may well not work since it may be evaluated
too early.
panel.last an expression to be evaluated after plotting has taken place but before the axes,
title and box are added. See the comments about panel.first.
asp the y/x aspect ratio

Details
Commonly used graphical parameters are:
 col: The colors for lines and points. Multiple colors can be specified so that each
point can be given its own color. If there are fewer colors than points they are
recycled in the standard fashion. Lines will all be plotted in the first colour
specified.
 bg: a vector of background colors for open plot symbols.
 pch: a vector of plotting characters or symbols.
 cex: a numerical vector giving the amount by which plotting characters and
symbols should be scaled relative to the default. This works as a multiple of
par("cex"). NULL and NA are equivalent to 1.0. Note that this does not affect
annotation.
 lty: the line type.
 lwd: the line width.

Examples
Speed <- cars$speed
Distance <- cars$dist
plot(Speed, Distance, panel.first = grid(8,8),pch = 0, cex = 1.2, col = "blue")
plot(Speed, Distance,panel.first = lines(stats::lowess(Speed, Distance), lty = "dashed"),pch
= 0, cex = 1.2, col = "blue")

23
 Graphs and Charts in R

3. BAR CHARTS

A bar chart is constructed to show frequencies of different categories of a categorical

variable in the given data. The horizontal axis represents the different categories and
verticle axis is used to show the number of cases (frequencies) of the characteristic in
which we are interested. We explain the construction of simple bar chart using R-
commands with example. To draw bar (or column) charts in R, use the barplot function.

Description
Creates a bar plot with vertical or horizontal bars.
Usage
barplot(height, ...)

## Default method:

barplot(height, width = 1, space = NULL,names.arg = NULL, legend.text = NULL,

beside = FALSE,horiz = FALSE, density = NULL, angle = 45,col = NULL,
border = par("fg"),main = NULL, sub = NULL, xlab = NULL, ylab = NULL,xlim
= NULL, ylim = NULL, xpd = TRUE, log = "",axes = TRUE, axisnames =
TRUE,cex.axis = par("cex.axis"), cex.names = par("cex.axis"),inside = TRUE,
plot = TRUE, axis.lty = 0, offset = 0,add = FALSE, args.legend = NULL, ...)

Arguments
height either a vector or matrix of values describing the bars which make up the plot.
If height is a vector, the plot consists of a sequence of rectangular bars with
heights given by the values in the vector. If height is a matrix and beside is
FALSE then each bar of the plot corresponds to a column of height, with the
values in the column giving the heights of stacked sub-bars making up the bar.
If height is a matrix and beside is TRUE, then the values in each column are
juxtaposed rather than stacked.
width optional vector of bar widths. Re-cycled to length the number of bars drawn.
Specifying a single value will have no visible effect unless xlim is specified.
space the amount of space (as a fraction of the average bar width) left before each
bar. May be given as a single number or one number per bar. If height is a
matrix and beside is TRUE, space may be specified by two numbers, where
the first is the space between bars in the same group, and the second the space
between the groups. If not given explicitly, it defaults to c(0,1) if height is a
matrix and beside is TRUE, and to 0.2 otherwise.
names.arg a vector of names to be plotted below each bar or group of bars. If this
argument is omitted, then the names are taken from the names attribute of
height if this is a vector, or the column names if it is a matrix.
legend.text a vector of text used to construct a legend for the plot, or a logical indicating
whether a legend should be included. This is only useful when height is a
matrix. In that case given legend labels should correspond to the rows of
height; if legend.text is true, the row names of height will be used as labels if
they are non-null.
beside a logical value. If FALSE, the columns of height are portrayed as stacked bars,
and if TRUE the columns are portrayed as juxtaposed bars.
 Graphs and Charts in R

horiz a logical value. If FALSE, the bars are drawn vertically with the first bar to the
left. If TRUE, the bars are drawn horizontally with the first at the bottom.
density a vector giving the density of shading lines, in lines per inch, for the bars or
bar components. The default value of NULL means that no shading lines are
drawn. Non-positive values of density also inhibit the drawing of shading
lines.
angle the slope of shading lines, given as an angle in degrees (counter-clockwise),
for the bars or bar components.
col a vector of colors for the bars or bar components. By default, grey is used if
height is a vector, and a gamma-corrected grey palette if height is a matrix.
border thecolor to be used for the border of the bars. Use border = NA to omit
borders. If there are shading lines, border = TRUE means use the same colour
for the border as for the shading lines.
main,sub overall and sub title for the plot.
xlab a label for the x axis.
ylab a label for the y axis.
xlim limits for the x axis.
ylim limits for the y axis.
xpd logical. Should bars be allowed to go outside region?
log string specifying if axis scales should be logarithmic; see plot.default.
axes logical. If TRUE, a vertical (or horizontal, if horiz is true) axis is drawn.
axisnames logical. If TRUE, and if there are names.arg (see above), the other axis is
drawn (with lty=0) and labeled.
cex.axis expansion factor for numeric axis labels.
cex.names expansion factor for axis names (bar labels).
inside logical. If TRUE, the lines which divide adjacent (non-stacked!) bars will be
drawn. Only applies when space = 0 (which it partly is when beside = TRUE).

Example 1: Annual sales (in lakh of Rs.) of a pharmaceutical firm for six years (1995-
2000) given below:

Table: Annual sales of pharmaceutical company

Year 1995 1996 1997 1998 1999 2000

Annual sales 15.0 25.0 27.0 28.0 26.0 26.6

Represent the data by bar chart.

Solution: Following are the appropriate R commands:
c1<-1995:2000; #Vector of years
c2<-c(15, 25, 27, 28, 26, 26.6); #Vector of sales
sales.year<-data.frame(year=c1, sales=c2);
sales.year; #Print table
attach (sales.year);
barplot (sales, xlab="Year", ylab ="Sales", main="", col="white");

25
 Graphs and Charts in R

We have used barplot function to create bar plot.

Figure: Bar diagram

Example 2: A group of 25 people was surveyed to find their soft-drink preference. The
categories of soft-drink used in the survey were Limca, Coca-cola, Pepsi and Mangola.
The data are:
3, 4, 1, 1, 3, 4, 3, 3, 1, 3, 2, 1, 2, 1, 2, 3, 2, 3, 1, 1, 1, 1, 4, 3, 1.

Represent the data by: (i) barplot of frequencies and (ii) barplot of proportions.

Solution: First we use the scan function to input data.

soft<-scan()
1:3 4 1 1 3 4 3 3 1 3 2 1 2 1 2 3 2 3 1 1 1 1 4 3 1
Read 25 items
barplot(table(soft), xlab="Soft-drink", ylab="Frequency of preferences", main="",
col="white");

Figure: Bar diagram with frequencies

Remark: Observe the first argument of barplot. One has to call barplot with summarized
data as its first argument; table(soft) command summarizes the data according to four
categoies of soft drinks. Then this summarized categorical data is represented in form of a
bar plot. Use the table function and see how you get frequencies of the four categories.
Note that barplot(soft) is not the correct way to obtain the desired bar plot. We must use
the following command to construct barplot of proportions:
 Graphs and Charts in R

barplot(table(soft)/length(soft), xlab="Soft-drink", ylab="Proportion of preferences",

main="", col="gray70");

Figure: Bar diagram of proportions

Remark: Observe the first argument of barplot function. It is now table(soft)/length(soft).
Here, length() function computes the number of observations in the data set.
Thustable(soft)/length(soft) computes the category wise proportions of the four soft
drinks.

4. SUBDIVIDED BAR DIAGRAM

Sometimes it is necessary to show break up of one variable into several components. A

simple bar diagram is not useful in such a situation; a subdivided or component bar
diagram may be used to represent data relating to such a variable over time or space. In
case of simple bar diagram, bars (or rectangles) are drawn to represent total magnitude of
the variable; each bar representing one category. In a subdivided bar plot, this bar is
divided into segments, each segment representing a component of the corresponding
category.

Example: Table shows the number of students in different faculties of a university.

Sr. No. Year Humanity Science Commerce

1 1996 2810 890 540
2 1997 3542 1363 471
3 1998 4301 1662 652
4 1999 5362 2071 895
5 2000 6593 2752 1113

(i) Represent the total number of students for different years by means of a simple bar
diagram.
(ii) Represent the data as a subdivided barplot.

27
 Graphs and Charts in R

Solution:
year.stud<-
matrix(c(2810,890,540,3542,1363,471,4301,1663,652,5362,2071,895,6593,2752,1113),
byrow=T, ncol=3);
rownames(year.stud)<-c("1996","1997","1998","1999","2000");
colnames(year.stud)<-c("Humanity","Science","Commerce");
year.stud;

Humanity Science Commerce

1996 2810 890 540
1997 3542 1363 471
1998 4301 1663 652
1999 5362 2071 895
2000 6593 2752 1113

total.stud<-margin.table(year.stud,1); # To prepare table of marginal totals we use

margin.table function. The second argument 1 indicates that we are taking row totals (for
column totals, the argument will be 2).
total.stud; #View the table of marginals.

1996 1997 1998 1999 2000

4240 5376 6616 8328 10458

barplot(total.stud, col="white", main="");

Figure: Bar diagram of marginals

barplot(t(year.stud), col=c("white", "gray80", "black"), main="", beside=FALSE);

 Graphs and Charts in R

Figure: Subdivided bar plot

legend(locator(1), legend=c("hu","sc","com"), fill=c("white", "gray80", "black"),
horiz=T);

Figure: Subdivided bar plot

Note:
(i) Observe that we have used matrix() function to input data in a table form.
(ii) The function barplot accepts a matrix as an argument; in this case, it is the
transpose of “year.stud”. (t(year.stud) means transpose of “year.stud”). The
default is a subdivided chart (that is why the argument “beside=FALSE”).
(iii)The locator function will allow you to choose a location for legend. The argument
1 means you can locate one point. The argument legend in legend function is
the text for legend and argument fill specifies the colours of the boxes next to
legend text. The argument horiz causes legend to be placed horizontally when
true. The default is vertical.

5. MULTIPLE BAR DIAGRAM

A multiple bar diagram is used for two or three-dimensional comparison. For comparison
of magnitudes of one variable in two or three aspects or comparison of magnitudes of two
or three variables, rectangles in a group are placed side by side. The R-commands are
similar to those used for subdivided barplot. Only we change the default value of argument
beside to true. We illustrate the construction of multiple barplot using the following data.

Example: The number of blood donation in the years 1995 and 2000 in various blood
groups are as given in the following table. Represent the data by multiple bar diagram.

29
 Graphs and Charts in R

Year Blood Groups

O A B AB
1995 1154 526 775 155
2000 1700 1125 1280 560

Solution:
donation<- matrix(c(1154, 1700, 526, 1125, 775, 1280, 155, 560), ncol=4);
colnames(donation) <-c("O", "A", "B", "AB");
rownames(donation) <-c("1995", "2000");
donation; #Print table

O A B AB
1995 1154 526 775 155
2000 1700 1125 1280 560
barplot(t(donation),col=c("white", "gray50", "gray70", "black"),main="",beside=T);
legend(locator(1), legend = c("O", "A", "B", "AB"), fill = c("white", "gray50", "gray70",
"black"), cex = 0.5);

Figure: Multiple barplot

6. PIE CHARTS

The circle is used in a pie chart to represent the whole, and “slices” are used to represent
the categories, one slice for each category. The size of a slice is proportional to relative
frequency of the corresponding category.

Example: The tax revenue of India (in crores of Rs.), as provided in 1984085 budget,
when broken into various sources are given below. Represent the data by a pie chart.

Sources Excise Customs Corporation Income tax Other

tax
Tax revenue 6526 7108 2568 560 763

Solution: R-commands for the pie chart are as follows:

pie.tax<- c(6526, 7108, 2568, 560, 763);
names(pie.tax) <- c("Excise", "Customs", "Corporation", "Income", "Other");
 Graphs and Charts in R

pie(pie.tax, main = "The Tax Revenue of India (1984-85)", col = c("white", "black",
"gray80", "gray50", "gray70"));

Figure: Pie plot

7. STEM-AND-LEAF PLOT

First, we split each number into two parts. The first part of the number serves as s “stem”
and the other part that follows is a “leaf”. We divide the range into intervals of fixed
length depending on the nature of data. These intervals are usually units of 0.5, 1 or 2
times a suitable power of 10. Further, the leaves belonging to each stem are ordered.

Example: Following are the numbers of rooms occupied in a hotel for 30 days. Draw
stem-and-leaf diagram.

20 14 21 29 43 17 15 26 8 14 39 23 16 46 28 11 26 35 26 28 30 22 23 7 32 19 22 18 27 9

Solution:
x <-scan()
1: 20 14 21 29 43 17 15 26 8 14 39 23 16 46 28 11 26 35 26 28 30 22 23 7 32 19 22 18 27
9
31:
Read 30 items
>stem(x); #We use stem function to draw stem-and-leaf diagram. Output of the command
is:

The decimal point is 1 digit(s) to the right of the |

0 | 789
1 | 144
1 | 56789
2 | 012233
2 | 6667889
3 | 02
3 | 59

31
 Graphs and Charts in R

4|3
4|6

Interpretation of stem and leaf plot:

(i) Each number is a two-digit number. The first digit constitutes the stem and the
second, a leaf. For instance, 2 is a stem and 0 is the leaf for first number 20.
Altogether, there are five stems, namely 0, 1, 2, 3 and 4, as the data ranges from 07
to 46. The number 7 has stem 0 and its leaf part is 7.
(ii) In this case, the fixed length of interval is (0.5*10) = 5. The stems are written to the
left of vertical line and corresponding leaves to the right.
(iii) From the diagram, reading horizontally along stem 1 we see that 11, 14 and 14
rooms are occupied on three days.
(iv) Given stem-and-leaf plot of two digit numbers we can reconstruct the given data.
(v) We can see that there is more concentration at the lower values.

Transformation or re-expression of the data: It is one of the most powerful tools

available to the data analyst. If used effectively, it enables one to make clearer
comparisons. The most common motive for the transformation is to arrive at symmetric
distribution of the data. Let us see how stem-and-leaf plot displays the effect of a
transformation with help of the following example.

Example: Below are 72 numbers, representing the counts of insects in agricultural

experimental unit treated with insecticides.

10 11 00 03 03 11 07 07 01 05 05 09 20 21 07 12 03 15 14 11 02 06 05 22 14 16 03 04 03
15 12 14 01 03 06 16 10 17 02 05 01 13 23 17 01 05 03 10 17 19 03 05 02 26 20 21 00 05
06 26 14 07 01 02 01 24 13 13 04 04 04 13

(i) Prepare a stem-and-leaf plot.

(ii) Consider the square root transformation. For the transformed data, prepare a stem-
and-leaf plot.
(iii)Comment on the result.

Solution: We enter data using the scan() function

x <- scan()
1: 10 11 00 03 03 11 07 07 01 05 05 09 20 21 07 12 03 15 14 11 02 06 05 22 14 16 03 04
03 15 12 14 01 03 06 16 10 17 02 05 01 13 23 17 01 05 03 10 17 19 03 05 02 26 20 21 00
05 06 26 14 07 01 02 01 24 13 13 04 04 04 13
73:
Read 72 items
>stem(x)

The decimal point is 1 digit(s) to the right of the |

0 | 001111112222333333334444
0 | 555555566677779
1 | 0001112233334444
1 | 55667779
2 | 0011234
 Graphs and Charts in R

2 | 66

The stem-and-leaf display of the data (x), output of the command stem(x), illustrates that
there is more concentration at lower values. In other words, the data display a positive
skew pattern. Now we give the command:

y=x^0.5; #Transform the data by using square root transformation.

stem(y); # Prepare stem-and-leaf for transformed data.

The output is:

The decimal point is at the |

0 | 00
1 | 000000444477777777
2 | 000022222224446666
3 | 0222333556666777799
4 | 0011145566789
5 | 11

We have considered square root transformation because it is commonly used for count
data. The stem-and-leaf display of the transformed data (y), is nearly symmetric.

8. FREQUENCY DISTRIBUTION

One way to achieve summary of the data is by forming a frequency table or a frequency
distribution. A frequency distribution is a list of all distinct data values along with the
number of times they occur.

Preparation of frequency table: We illustrate the construction of frequency table with help
of the following example.

Simulate 100 rolls of a die and prepare a frequency table.

roll<- sample(1:6, 100, replace = T);
roll; # Print the simulated sample.

[1] 5 5 4 3 4 6 1 2 5 2 5 6 1 2 1 6 4 2 4 5 4 4 3 6 5 4 2 3 5 4 6 2 6 3 2 2 2
[38] 4 4 3 5 3 2 4 1 5 2 5 3 1 6 3 2 1 2 5 2 3 1 2 2 4 3 4 3 1 6 2 6 5 3 6 3 5
[75] 1 5 3 3 3 5 5 2 2 4 1 3 6 6 6 3 6 6 6 6 3 5 2 4 5 4

rolltab<- table(roll); #Prepare frequency table.

rolltab; #Print table.

Following is the formatted output:

roll
1 2 3 4 5 6
10 20 19 16 18 17

33
 Graphs and Charts in R

Note: With the “sample” command, every time we will get a different sample and hence a
different frequency table.

Preparation of frequency distribution: We illustrate the construction of frequency

distribution with the help of following example.

Following are scores of eighty students in the examination:

62 73 85 42 68 54 38 27 32 63 68 69 75 59 52 58 36 85 88 72 52 52 63 68 29 73 29 76 29
57 46 43 28 32 09 66 72 68 42 76 38 38 39 28 19 12 78 72 92 82 72 33 92 69 28 39 85 59
68 52 85 59 76 80 72 74 54 48 29 36 10 82 58 88 68 58 46 37 29 35

score<- scan()

1: 62 73 85 42 68 54 38 27 32 63 68 69 75 59 52 58 36 85 88 72 52 52 63 68 29 73 29 76
29 57 46 43 28 32 09 66 72 68 42 76 38 38 39 28 19 12 78 72 92 82 72 33 92 69 28 39 85
59 68 52 85 59 76 80 72 74 54 48 29 36 10 82 58 88 68 58 46 37 29 35
81:
Read 80 items
n <- length(score);
k <-6; # Number of classes
w <- 15; # Width of class interval
i <-1:6;
x <-c(12.5+w*(i-1)); #12.5 is the midpoint of the first class, namely, 5-19
x; # Print x
[1] 12.5 27.5 42.5 57.5 72.5 87.5
f <-1:6;for(i in 1:6){f[i]<-length(score[(x[i]-w/2)<=score & score <=(x[i]+w/2)])};
f; #Print vector frequencies
[1] 4 13 15 16 23 11
fr.dist<-data.frame(midvalue = x, frequency = f);
fr.dist; # Print frequency distribution

midvalue frequency
1 12.5 4
2 27.5 13
3 42.5 15
4 57.5 16
5 72.5 23
6 87.5 11

Example: Following are heights of 45 students. Heights are recorded in centimetres.

Prepare a frequency distribution of the data. Use the rule: First k for which 2k>=n, to
determine the number of k classes for n = 45 observations.

170 151 154 160 158 154 171 156 160 157 160 157 148 165 158 159 155 151 152 161
156 164 156 163 174 153 170 149 166 154 166 160 160 161 154 163 164 160 148 162
167 165 158 158 176

Solution: From the rule we get k = 6. The smallest observation is 148 and the largest
observation is 176. The difference is 28. The class width is 28/6=6.67. This number is
 Graphs and Charts in R

nearly equal to 5, which is more convenient to use. Height is a continuous variable. What
needs to be done is to choose a set of proper class intervals or class boundaries. For
example, with a 5 centimetres class interval, all heights between and including 148 and
152 will be placed in same class. Since these heights are measured to the nearest
centimetre, this particular class interval would extend from 147.5 to 152.5. These limits
are known as proper class limits or class boundaries. These class boundaries will not
change the mid-value of the class. The mid-value of the first class is 150.

height<- scan()
1: 170 151 154 160 158 154 171 156 160 157 160 157 148 165 158 159 155 151 152 161
156 164 156 163 174 153 170 149 166 154 166 160 160 161 154 163 164 160 148 162
167 165 158 158 176
46:
Read 45 items
n <- length(height);
k <- 6;
w <-5;
i <-1:6;
x <-c(150+w*(i-1)); # Vector of mid points
x;
The output is:
[1] 150 155 160 165 170 175

Following loop will compute the frequencies of the classes

f <-1:6; for(i in 1:6){f[i] <- length(height[(x[i]/w/2) <= height & height <= (x[i]+w/2)])};
f;
[1] 6 17 31 40 43 45

fr.dist<- data.frame(midvalue = x, frequency = f);

fr.dist;

midvalue frequency
1 150 6
2 155 17
3 160 31
4 165 40
5 170 43
6 175 45

Graphical representation of frequency distribution: Frequency distributions are easier

to visualize if they are represented graphically. We illustrate the construction of the
various graphs to represent frequency distribution using R-commands.

Rod or spike graph: The graph is used to represent the frequency table or ungrouped
frequency distribution of a discrete variable.
Example: Four similar coins are tossed 100 times. The number of heads (x) in each of the
100 tosses was noticed. Following is the frequency table of number of heads in 100 tosses
of the four coins.

35
 Graphs and Charts in R

Number of heads 0 1 2 3 4
(x)
Frequency (f) 6 28 36 25 5

Prepare a rod chart or spike chart.

Solution: R-commands for spike chart are as follows.

X <- c(0, 1, 2, 3, 4);
f <- c(6, 28, 36, 25, 5);
plot(x, f, "h", xlab="Number of success", ylab="Frequency");
points(x, f, pch=16); # The argument pch gives the type of the point

Figure: Spike chart

9. HISTOGRAM

Histogram is commonly used for continuous variables. It may also be used for a grouped
frequency distribution of discrete variable. It is drawn as described below.

(a) The class boundaries are marked along the x-axis and are taken as bases of the
rectangles.
(b) On each base (class), a rectangle is drawn with the area proportional to the
frequency or relative frequency (that is, proportion) of that class.

Illustration: Suppose we are given 25 observations as follows:

29.6 28.2 19.6 13.7 13.0 7.8 3.4 2.0 1.9 1.0 0.7 0.4 0.4 0.3 0.3 0.3 0.3 0.3 0.2 0.2 0.1 0.1
0.1 0.1 0.1

First we use the scan function to input the data and then draw some histograms.

x <- scan()
1: 29.6 28.2 19.6 13.7 13.0 7.8 3.4 2.0 1.9 1.0 0.7 0.4 0.4 0.3 0.3 0.3 0.3 0.3 0.2 0.2 0.1 0.1
0.1 0.1 0.1
26:
 Graphs and Charts in R

Read 25 items
hist(x, xlab ="", ylab="Frequency", main="", col=gray(0.9));

Figure: Histogram of frequencies

hist(x, prob=T, xlab="", ylab="Relative Frequency", main="", col=gray(0.9)); #Histogram

of proportions

Figure: Histogram of proportions

hist(x, breaks=c(0, 1, 2, 3, 4, 5, 10, 20, max(x))); #Specify break point i.e. class limits. 9
break points are specified.

Figure: Histogram of proportions with eight break points

37
 Graphs and Charts in R

hist(x, breaks=c(0, 5, 10, 20,max(x))); #Only 5 break points are specified.

Figure: Histogram of proportions with eight break points

Histogram for grouped frequency distribution: The histogram drawn so far used the
raw data as their argument. It is possible that we are given data in the form of a grouped
frequency distribution. The following example explains the method of drawing histogram
for such a data.

Illustration: Example: Following are heights of 45 students. Heights are recorded in

centimetres.

170 151 154 160 158 154 171 156 160 157 160 157 148 165 158 159 155 151 152 161
156 164 156 163 174 153 170 149 166 154 166 160 160 161 154 163 164 160 148 162
167 165 158 158 176

A. Suppose we are given raw data:

B. Suppose we given grouped frequency distribution constructed from the raw data of part
A

Midvalue Frequency
1 150 6
2 155 17
3 160 31
4 165 40
5 170 43
6 175 45

R-commands:

midvalue<- c(150, 155, 160, 165, 170, 175)

frequency<- c(6, 11, 14, 9, 3, 2);
y <- rep(midvalue, frequency);
brk<- seq(147.5, 177.5, 5);
brk;
[1] 147.5 152.5 157.5 162.5 167.5 172.5 177.5
 Graphs and Charts in R

Par(mfrow=c(1, 2));
hist(x, breaks=brk, col="white");
hist(y, breaks=brk, col="white");
par(mfrow=c(1, 1));

Figure: Histograms for Raw (x) and grouped data

Note: Observe that vector x contains raw data i.e. 45 observations. The vector y also
contains 45 observations generated from given grouped frequency distribution under the
assumption that the frequency of each class is concentrated at the mid point of the class
interval. It may be noted that the breakpoints for histogram of raw data are taken as class
limits of the grouped data. We have prepared histogram for x as well as y for comparison.
Thus, histogram for frequency distribution can be constructed by using the same class
limits of the frequency distribution as the break points and making usual assumption that
the frequency of the class is concentrated at mid points of the class intervals.

Example: The following table shows the frequency distribution for Book Store Sales
Data. Construct the histogram for this data set.

Table: Frequency distribution of Daily sales

Class Interval Frequency

(Daily sales
0-25 5
25-50 8
50-75 13
75-100 11
100-125 3

Solution: The following figure displays the histogram using the following R-commands.

midx<- seq(12.5, 112.5, 25);

frequency<- c(5, 8, 13, 11, 3);
y <-rep(midx, frequency);
brk<- seq(0, 125, 25);
hist(y, breaks = brk, xlab = "Sales", main = "", col = "gray70");

39
 Graphs and Charts in R

Histogram of Sales of Books

Remarks: The rule of equal class intervals is inconvenient when data are spread over wide
range but are highly concentrated in a small part of the range with relatively few numbers
elsewhere. Using smaller intervals where the data are highly concentrated and larger
intervals where the data are sparse helps to reduce the loss of information due to grouping.
Tabulations of income, population and some such characteristics in official reports are
often made with unequal class intervals. We consider one such example.

Example: The following table shows the projected population (in millions) of a country
for the year 2005. The projections are broken down by age groups where grouping follow
natural areas of interest such as preschool (below 5 years), education group ( divided into
3 intervals, 5-13, 14-17 and 18-24), adult group (covering 25-64 years with 4 intervals of
equal widths) and finally senior citizens’ group (65 and above). Construct a histogram for
the data.

Table: Projected population

Age Group Projected Population

Below 5 18
5-14 35
14-18 16
18-25 25
25-35 34
35-45 41
45-55 36
55-65 22
65 and above 32

Solution: We assume arbitrarily large upper bound, say 100, for the last class, which is
open end class.
midx<- seq(12.5, 112.5, 25);
frequency<- c(5, 8, 13, 11, 3);
 Graphs and Charts in R

y <-rep(midx, frequency);
brk<- seq(0, 125, 25);
hist(y, breaks = brk, xlab = "Sales", main = "", col = "gray70");
midx<-c(2.5, 9.5, 16, 25, 30, 40, 50, 60, 82.5);
frequency<-c(18, 35, 16, 25, 34, 41, 36, 22, 32);
brk<-c(0.5, 14, 18, 25, 35, 45, 55, 65, 100);
y <-rep(midx, frequency);
hist(y, breaks=brk, xlab="Age Group", ylab="Agewise Projected Population",
col="gray70");

Figure: Histogram of frequency distribution with open-end-class

10. FREQUENCY POLYGON

One graphical method to clarify and amplify the content of the frequency table if the
frequency polygon. It is obtained by joining the points , where is the midpoint of
th
the i class interval and is the corresponding frequency. In order to close the frequency
polygon, two additional points and are added. Either (i) is the lower
limit of the lowest class interval and is the upper limit of the highest class interval or
(ii) is the midpoint of the class interval previous to the lowest class interval and is
the midpoint of the class interval next to the highest class interval. We explain R-
commands for plotting frequency polygon with examples.

Example: Following table shows the frequency distribution of college student according to
their pocket money (daily). Draw histogram and frequency polygon on the same graph
paper.

Frequency Distribution of Pocket Money

Pocket Money (In Rs) 20-29 30-39 40- 50-59 60-69 70-79 80-89
49
No. of Students 10 24 18 12 8 5 3

Solution: Figure shows the graph drawn using following R-commands.

midx<- seq(25, 85, 10);
frequency<-c(10, 24, 18, 12, 8, 5, 3);

41
 Graphs and Charts in R

x<- rep(midx, frequency);

brk<-seq(20, 90, 10);
hist(x, breaks=brk, main="", xlab="Pocket Money", ylab="No. of Students");
temp<-hist(x, xlab="Pocket Money", ylab="No. of Students", main="", col=gray(0.8));
lines(c(min(temp$breaks), temp$mids, max(temp$breaks)), lwd = 2, c(0, temp$counts, 0),
type = "l");

Figure: Histogram and frequency polygon

Example: Following table gives the distribution of deaths from scarlet fever classified
according to age. Plot frequency polygon.

Frequency Distribution of Deaths

Age (mid value) 2.5 7.5 12.5 17.5 22.5 27.5 32.5
No. of Deaths 322 213 70 27 26 17 16
Age (mid value) 37.5 42.5 47.5 52.5 57.5 62.5
No. of Deaths 11 10 7 6 5 1

Solution: R-commands:
midx<- seq(-2.5, 67.5, 5);
frequency<-c(0, 322, 213, 70, 27, 26, 17, 16, 11, 10, 7, 6, 5, 1, 0);
plot(midx, frequency, type="o", xlab = "Age", ylab = "Deaths", main="", lwd = 2);

Figure: Age-wise deaths from scarlet fever

 Graphs and Charts in R

Note: We have used the plot() function to draw frequency polygon. The first two
arguments to this function are x and y coordinates respectively. The third argument
specifies the type of plotting parameter. The default for this is “p” (point. You can use “l”
(line), or “b” (line and point both) to see the difference between the outputs. We have used
“o” (overlap of point and line).

Graphs showing cumulative frequencies: “Less than” type cumulative frequency of a

class is the number of observations less than or equal to the upper limit (upper class
boundary) of the class. Similarly, “more than” type cumulative frequency of a class is the
number of observations bigger than lower limit (lower class boundary) of the class. The
cumulative frequency distribution is represented by “ogive” or cumulative frequency
curve. We illustrate the construction of these two types of curves.

Example: The following frequency distribution relates to lives of 400 light bulbs. Draw
“less than” and “more than” type ogive curves on the small graph paper.

Table: Life of Bulbs

Life of bulb (in hrs) 600-699 700-799 800-899 900-999 1000-

1099
No. of bulbs 85 77 124 78 36

Solution: The following figure is obtained by using following R-commands:

f <-c(0, 85, 77, 124, 78, 36, 0);
lc<-cumsum(f);
lc;
[1] 0 85 162 286 364 400 400
uc<-1:7;for(i in 7:1){uc[i]<-sum(f[7:i])};uc
[1] 400 400 315 238 114 36 0
lbx<-seq(499.5, 1099.5, 100); #The vector of lower class boundaries.
lbx;
[1] 499.5 599.5 699.5 799.5 899.5 999.5 1099.5
ubx<-seq(599.5, 1199.5, 100); #The vector of upper class boundaries.
ubx:
[1] 599.5 699.5 799.5 899.5 999.5 1099.5 1199.5
plot(ubx, lc, type = "l", xlim=c(499.5, 1199.5), xlab = "Class interval", ylab="Cumulative
Frequency", main= "", lwd = 2);
lines(lbx, uc, lty = 2, lwd = 2);

43
 Graphs and Charts in R

Figure: Ogive

Note: We have used one more argument xlim (limits of x) to plot function. It is necessary
because we want to include lower as well as upper class boundaries.

REFERENCES

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

Purohit, G. S., Gore S. D. and Deshmukh, S. R. (2010). Statistics using R. Narosa
Publishing House Pvt Ltd., New Delhi.