2.

7 Diagrammatic representation of data

Graphs, charts, maps, pictures, etc. are attractive and effective means for presentation of
statistical data. Diagrams are readily capuble of revealing some features of the exhibited data.
It should be noted that the selection of the
of the
appropriate diagram depends mainly on the nature
given data.
Following are the important merits and demerits of the diagrammatic mode of presenting data.
Merits
(i) It is simple understand even by laymen.
to
(ii) It is very essential for conveying statistical information to the general public in a short
time.
(ii) In this approach, one may acquire idea regarding the
data at a
some significance of the presented
glance.
(iv) This mode is capable of creating lasting impression.
(v) Two or more series of data can easily be compared.
Demerts
(i) Diagrams fail to represent details; they only show the
general nature of the given data.
(ii) Usually, a diagram represents the figures in
of the cases, precision of data has to be approximate form and, as such, in most
sacrificed.
(iii) Construction of a diagram requires sufficient time.
(iv) Only limited information can be presented in a diagram.
Our subsequent discussion will relate to some of the
used commonly diagrams.
1. Line diagram This diagram is meant for representing
:
exhibits the relationship of the variable under chronological data. In fact,it
variable (e.g. sales of coffee of a company)
study with time. The successive values of the
may be specified for
(called point data) or for different periods of time (referred to as individual points of time
period data).
 In constructing a line diagram, two axes of co-ordinates are taken, the horizontal one for
time and the vertical one for the variable. The scale for each axis is then selected and the
data are ploted as different points on the plane, the plotting of variable values being done
against the points of time (for point data) or the mid-points of the time intervals (for period
data). The successive points are now joined by straight line segments and the chart so obtained
is called a line diagram for the given data.
It should be noted that, in the chart, none of the axes should be very long or very snOt
relation to the other; otherwise, the fluctuations in the values of the variable would be over-
emphasized (when horizontal axis is shorter) or almost ironed out (when vertical axis is shorter).
Again, the zero of the vertical scale should be shown in the diagram to avoid wrong impression.
In doing so, we shall find the chart far away from the horizontal axis when the actual range
of variation of the values is much different from zero. Here the chart may be brought down
and made agreeable to the eye by showing a definite break in the vertical axis.

Two or more mutually related time series data having same unit of measurement can be

represented using the same axes of coordinates, by drawing a number of line diagrams, one
for each series. These different line diagrams are usually distinguished by using distinct pattem
of lines such as broken lines, continuous lines, etc.
It should be noted that a line diagram shows the absolute change in the value of the variable
with the change in time. If the given variable changes by a constant rate, then the diagram
will be exactly a straight line.

1llustration 2.1
line diagram.
Represent the following data by a

1961 1966 1971 1976 1981

Year
12 15 21.2 20 27.5
Population (in lakhs)
30

25
20

15

10

1971 1976 1981

1961 1966
Year>
from 1961 to 1981.
of a region during the period
the population
Chart 2.1 Line diagram showing
 2. Ratio chart: It is a special kind ofline diagram that shows the relative change in the
value of the given variable with
respect to change in time.
The notable feature of this chart is that the horizontal scale is of arithmeti (or natural)

type but the vertical scale is logarithmic, In a ratio chart, it is evident that equal distances in
the natural scale indicate equal differences, while they represent equal relatives in the other

scale. The distance between any two values on the vertical scale of this chart is proportional
to the difference of their logarithms.
In a ratio chart (also called a semi-logarithmic chart), vertical rulings are equally apart but
horizontal rulings are not so. Such a chart is used when one wants to show the relative changes
in the variable rather than the absolute changes. If the values of the variable change at a
constant ratio, then the ratio chart becomes a perfect straight line.
one may draw
For comparison of the ratesof relative change of different series of data,
ratio charts for the series and compare their slopes. For this,
the different series are not required
to be in the same unit.
data having large difference in the
A ratio chart also enables one to exhibit chronological
values of the variable under study.

lustration2.2
in certain industrial factory are given below:
Accounts of gains
1970 1971 1972 1973 1974 1975 1976
Year
Gain (Rs. crores): 3.5 5.35 6.75 8.2 11.4 19.2 31.4
Draw ratio chart on a natural scale graph paper.

1.6

14

5 1.2

1.0
0.8
0.6t
0.4 1970 1971 1972 1973 1974 1975 1976
Years
Chart 2.2. Ratio chart showing the gain of a factory during 1970-1976.
3. Bar diagram : This mode of diagrammatic representation comprises a number of
some specific category
equidistant rectangles (termed as bars), each of them being meant for
the base
of the available data. Bars of common width (conveniently chosen) are drawn on
value of the corresponding category. Bars
line, the length or height of a bar representing the
or horizontal, depending on the nature of statistical data. Generally,
vertical
may be vertical
bars are drawn for time-series data and horizontal bars for data varying over space.

As already noted, bar diagrams are applicable to represent both time-series and spatial-
 series data. Besides, these diagrams are useful in exhibiting data classified on the basis of
quantitative characters and qualitative characters.

Illustration 2.3
The production of rubber in 1970 in a number of rubber-producing countries is as follows :

Country Production of Rubber ("000 tonnes)

Indonesia 786
Thailand 269
India 75
Malayasia 1240
Others 170
Give a suitable diagrammatic representation of the above data.
500 1000 1300

Indonesia

Thailand

India

Malayasia

Others

500 1000 1300

Production of rubber (' 000 tones)
Chart 2.3 Bar diagram showing the production of rubber in 1970 in some countries.
Multiple bar diagram is a variant of the bar diagram, which is used to compare two or
more series of data on the same variable or for showing different components of an item (such
asrural and urban population of a state recorded in a few censuses). In this diagram, several
sets of bars are drawn so that bars for a period or a related phenomenon are put together and
uniform gap is maintained between any two sets of bars.

Illustration 2.4
The percentage of literacy in West Bengal is shown below separately for males and females
for 4 years.
Years
1941 1951 1961 1971

Males 27.4 34.1 40.1 44.8

Females 3.9 12.3 17.0 22.1
 501

40 MALES
30 FEMALES
20

10|

1941 1951 1961 1971

Chart 2.4 Multiple bar diagram showing the percentage of literacy in W. Bengal for males and females
for 4 years.

Component bar diagrams used for comparing the

are magnitudes of different components
of an item among themselves and also with the total.
for several years. Then
Suppose we are given the population of a country, classified by sex,
we draw several bars, as in the case of bar diagram, to represent the total population in different
a line drawn parallel to the base so that
years. Next, each bar is divided into two parts by
heights of the parts are proportional to male and female populations of the corresponding period.
4. Divided bar diagram and Pie diagram : In some situations, the values of a variable are
the
available for a number of components
and
comparison among the different componcents or
relation between each part and the whole may be necessary. n this context, the proportions
or percentages of the various components are given more importance than the absolute values.
So we have to think of the diagrammatic representation of tbhe percentages for difterent
categories. Divided bar diagram is an appropriate device for this purpose. Here a single bar
of suitable length and width is taken and its areais considered as 100. The area of the bar is
hen divided into a number of parts, depending on the number of categories, with the help of
lines drawn parallel to the base, so that the area of a part represents the percentage for the

category concerned.
A pie diagram is another appropriate diagram used for exhibiting the relative sizes of the
different parts of a whole. In this case, a circleis partitionedinto several sectors by drawing
angles at the centre, the area of each sector indicating the corresponding percentage. n fact,
the area enclosed by the circle is regarded as 100. Since the total angle at the centre is 360°,
the desired angle for some particular category will be 3.6 times the relevant percentage. The
diagram thus constructed is termed as a pie diagram.

lustration 2.5
Represent the following production cost of sugar in a certain week under different heads
in a sugar factory by pie chart
 Source Production cost ('000 Rs.)
Raw material 192.0
Labour 153.6
Direct production 57.6 To Tal 49Po
Others 76.8
We first exhibit the percentages and corresponding angles for each of the given sources as
follows

TABLE 2.1
NECESSARY PERCENTAGES AND ANGLES FOR PIE CHART
Source Percentag8e Angle
x10o
480
=40 Raw material 40 144.0 40x360-14
53CYI0o/432 Labour 32 I15.20
5T8xB46- 1 32x360-115.2°
Direct production 12 43.20
480 Others 16 57.6 23o -43
Total 100 360.0 1305

Raw Material
(40%)

E Labour
(32%) Others

Direct HN
32
production
li
LMHTH
Chart 2.5 Pie chart showing the percentage of production cost
of sugar in a factory under different heads.

plvi dad b an dioqmom shrar