UNIT 15 GRAPHICAL PRESENTATION OF Graphical Presentation of

Data-I
DATA-I
Structure
15.1 Introduction
Objectives
15.2 Graphical Presentation
15.3 Types of Graphs
Histogram
Frequency Polygon
Frequency Curve
Ogive
15.4 Summary
15.5 Solutions/Answers

15.1 INTRODUCTION
An important function of Statistics is to present the complex and huge data in
such a way that they can easily understandable. In previous unit, we have
discussed the diagrammatic presentation of the data where we have become
familiar with some of the most commonly used diagrams. After discussing the
diagrammatic presentation of data, we are now moving towards the graphical
presentation of data. The graphs are plotted for frequency distributions and are
used to interpolate/extrapolate items in a series including locating various
partition values. In this unit, we shall discuss some of the most useful and
commonly used graphs.
The graphical presentation can be divided into two categories
(i) Graphs for frequency distributions.
(ii) Graphs for time series.
In this unit, we will concentrate ourselves to the graphs for frequency
distributions only. In this regard, we would like to discuss the most commonly
used graphs for frequency distributions, i.e. Histograms, Frequency polygon,
Frequency curve and Cumulative frequency curves, or Ogives.
Objectives
After studying this unit, you would be able to:
 describe the graphical presentation;
 explain the advantages of graphical presentation;
 draw the histogram for continuous frequency distribution;
 draw the frequency polygon for a frequency distribution;
 draw the frequency curves of different shapes; and
 draw the cumulative frequency curves.

47
Presentation of Data
15.2 GRAPHICAL PRESENTATION
A graphical presentation is a geometric image of a set of data. Graphical
presentation is done for both frequency distributions and times series. Unlike
diagrams, they are used to locate partition values like median, quartiles, etc, in
particular, and interpolate/extrapolate items in a series, in general. They are
also used to measure absolute as well as relative changes in the data. Another
important feature of graphs is that if a person once sees the graphs, the figure
representing the graphs is kept in his/her brain for a long time. They also help
us in studying cause and affect relationship between two variables. The graph
of a frequency distribution presents the huge data in an interesting and effective
manner and brings to light the salient features of the data at a glance. Before
closing this Sec. let us see some advantages of graphical presentation.
Advantages of Graphical Presentation
The following are some advantages of the graphical presentation:
 It simplifies the complexity of data and makes it readily understandable.
 It attracts attention of people.
 It saves time and efforts to understand the facts.
 It makes comparison easy.
 A graph describes the relationship between two or more variables.
After going through the advantages of graphical presentation of data, you were
keen to know the commonly used graphs to represent the data and how these
graphs are drawn. Next section will address these issues.

15.3 TYPES OF GRAPHS

Now days a large variety of graphs are in practical use. However, we shall
discuss only some important graphs which are more popularly used in practice.
The various types of graphs can be divided broadly under the following two
heads:
(i) Graphs of Frequency Distributions
(ii) Graphs for Time Series Data
In this Sec. we will focus on graphs for frequency distributions and graphs for
time series data will be discussed in next unit, i.e. Unit 16.
Graphs of Frequency Distributions
The graphical presentation of frequency distributions is drawn for discrete as
well as continuous frequency distributions.
Let us first consider the frequency distribution of a discrete variable.
To represent a discrete frequency distribution graphically, we take two
rectangular axes of co-ordinates, the horizontal axis for the variable and the
vertical axis for the frequency. The different values of the variable are then
located as points on the horizontal axis. At each of these points, a perpendicular
bar is drawn to present the corresponding frequency.
Such a diagram is called a ‘Frequency Bar Diagram’. For example, if we take
the frequency distribution for the number of peas per pod for 198 pods as given
in Table 15.1:

48
 Table 15.1 Graphical Presentation of
Data-I
No of peas per pod 1 2 3 4 5 6 7
Frequency (number of pods) 14 23 66 40 26 18 11

Then, the frequency bar diagram is shown in Fig. 15.1:

Fig. 15.1: Frequency Bar Diagram for the Frequency Distribution of Number of the Peas for 198 Pods.

Note 1: O represents origin and choice of scale used along horizontal and
vertical axes depends upon given data.
Now, we take the case of frequency distribution of a continuous variable.
The following are the most commonly used graphs for continuous frequency
distributions:
(i) Histogram
(ii) Frequency Polygon
(iii) Frequency Curve
(iv) Cumulative Frequency Curve or Ogives
Let us discuss these one by one:
(i) Histogram
In previous example, we have discussed how a graph is drawn for discrete
frequency distribution.
For the continuous frequency distribution, a better way to represent the data
graphically is to use a histogram. A histogram is drawn by constructing
adjacent rectangles over the class intervals such that the length of the
rectangles is proportional to the corresponding class-frequencies.
Histogram is similar to a bar diagram which represents a frequency distribution
with continuous classes. The width of all bars is equal to class interval. Each
rectangle is joined with the other so as to give a continuous picture.
The class-boundaries are located on the horizontal axis. If the class-intervals
are of equal size, the heights of the rectangles will be proportional to the class-
frequencies themselves. If the class-intervals are not of equal size, the heights
of the rectangles will be proportional to the ratios of the frequencies to the
49
Presentation of Data width of the corresponding classes. In other words, the frequencies of the class-
intervals having the least width are written as they are and the frequencies of
other class intervals are written as follows:
Given frequency
  The least width  … (15.1)
Width of its Class-interval
Let us draw a histogram to the following frequency distribution given below in
the table 15.2
Table 15.2
Class 0-10 10-20 20-30 30-40 40-50 50-60 60-70 70-80
Intervals
Frequency 2 3 13 18 9 7 6 2

Histogram for the above data is given below.

Fig. 15.2: Histogram for Frequency Distribution when Class-intervals are of Equal Width.

Now, let us consider the frequency distribution for unequal class intervals as
given in the Table 15.3
Table 15.3
Class 0-10 10-20 20-30 30-40 40-70 70-80 80-100
Frequency 20 32 8 2 60 35 10
As it is a case of unequal class intervals, so we have to adjust the frequencies of
the classes 40-70 and 80-100 by the formula suggested in equation 15.1. These
calculations are shown in table 15.4 given below:
Table 15.4
Class Interval Frequency Width of Heights of the rectangles
(CI) (CI)
0-10 20 10 20
10-20 32 10 32
20-30 8 10 8
30-40 2 10 2
40-70 60 30 (60/30)  10 = 20
70-80 35 10 35
80-100 10 20 (10/20)  10 = 5

50
The histogram for this frequency distribution is shown in Fig. 15.3. Graphical Presentation of
Data-I

Fig. 15.3: Histogram for Frequency Distribution when Class Intervals are of Unequal Width
Note 2: Sometimes, a histogram is also used for the frequency distribution of a
discrete variable. Each value of the discrete variable is regarded as the mid-
point of an interval. But generally, its use is not recommended, because in
discrete case each frequency actually corresponds to a single point and not to
an interval.
Now, you can try the following exercises.
E1) Draw a histogram from the following data
Class Interval: 0-10 10-20 20-30 30-40 40-50 50-60 60-70 70-80 80-90
Frequency: 3 5 10 14 24 17 14 10 3
E2) Draw a histogram for the following frequency distribution
Wage (Rs): 0-5 5-10 10-15 15-20 20-25 25-30 30-35 35-40 40-45 45-50
No. of
Workers: 30 70 100 110 140 150 130 100 90 60

(ii) Frequency Polygon

Another method of presenting a frequency distribution graphically other than
histogram is to use a frequency polygon. In order to draw the graph of a
frequency polygon, first of all the mid values of all the class intervals and the
corresponding frequencies are plotted as points with the help of the rectangular
co-ordinate axes. Secondly, we join these plotted points by line segments. The
graph thus obtained is known as frequency polygon, but one important point to
keep in mind is that whenever a frequency polygon is required we take two
imaginary class intervals each with frequency zero, one just before the first
class interval and other just after the last class interval. Addition of these two
class intervals facilitate the existence of the property that
Area under the polygon = Area of the histogram
For example, if we take the frequency distribution as given in Table 15.2 then,
we have to first plot the points (5, 2), (15, 3), …, (75, 2) on graph paper along
with the horizontal bars. Then we join the successive points (including the mid
points of two imaginary class intervals each with zero frequency) by line
segments to get a frequency polygon. The frequency polygon for frequency
distribution given in Table 15.2 is shown in Fig. 15.4.

51
Presentation of Data

Fig. 15.4: Frequency Polygon for the Frequency Distribution given in Table 15.2.

Note 3: In some cases first class interval does not start from zero. In such
situations we mark a kink on the horizontal axis, which will indicates the
continuity of the scale starting from zero. Let us take an example of this type.

Example 1: Draw a frequency polygon for the following frequency

distribution:

Class 40-50 50-60 60-70 70-80 80-90 90-100 100-110 110-120

Interval
Frequency 4 10 11 13 18 14 11 5

Solution: Frequency polygon for the given data is shown in Fig. 15.5:

Fig. 15.5: Frequency Polygon for Continuous Frequency Distribution.

Now, you can try the following exercise.

E3) Draw a frequency polygon from the following data
Wage: 0-5 5-10 10-15 15-20 20-25 25-30 30-35 35-40 40-45 45-50
No of
Workers: 20 50 80 110 140 150 120 150 100 80

(iii) Frequency Curve

In simple words frequency curve is a smooth curve obtained by joining the

52
points (not necessary all points) of the frequency polygon such that Graphical Presentation of
Data-I
(a) Like frequency curve it also starts from the base line (horizontal axis) and
ends at the base line.
(b) Area under frequency curve remains approximately equal to the area under
the frequency polygon.
In other words, let us try to explain the concept theoretically. Suppose we draw
a sample of size n from a large population. Frequency curve is the graph of a
continuous variable. So theoretically continuity of the variable implies that
whatever small class interval we take there will be some observations in that
class interval. That is, in this case there will be large number of line segments
and the frequency polygon tends to coincides with the smooth curve passing
through these points as sample size (n) increases. This smooth curve is known
as frequency curve.
In the following example we have drawn both frequency polygon and
frequency curve to make the idea clear for you.
Example 2: Draw frequency polygon and frequency curve for the following
frequency distribution.

Class 10-20 20-30 30-40 40-50 50-60 60-70 70-80 80-90

Intervals
frequency 2 5 8 15 18 10 3 1

Solution: Frequency polygon and frequency curve for the above data is given
below in Fig. 15.6.

Fig. 15.6: Frequency Curve along with Frequency Polygon.

On the next page some important types of frequency curves are given which
are generally obtained in the graphical presentations of frequency distributions.
That is, symmetrical, positively skewed, negatively skewed, J shaped, U
shaped, bimodal and multimodal frequency curves. You note that the shapes of
these curves justify their names.

53
Presentation of Data

Fig. 15.7

Now, you can try the following exercise.

E4) Draw a frequency curve from the following frequency distribution
Class Intervals: 1-2 2-3 3-4 4-5 5-6 6-7 7-8 8-9
Frequency: 3 9 20 25 18 12 6 3
(iv) Cumulative Frequency Curves
In sub Sec. 13.2.2 of the Unit 13 of this course you have already studied the
concept of cumulative frequency, cumulative frequency distribution, less than
cumulative frequency, more than cumulative frequency, less than cumulative
frequency distribution, more than cumulative frequency distribution, etc. Here
our aim is to study graphical presentation of less than and more than
cumulative frequency distributions, which are known as less than frequency
curve (or less than ogive) and more than frequency curve (or more than ogive)
respectively.
For drawing less than cumulative frequency curve (or less than ogive), first of
all the cumulative frequencies are plotted against the values (upper limits of the
class intervals) up to which they correspond and then we simply join the points
by line segments, curve thus obtained is known as less than ogive. Similarly,
more than frequency curve (more than ogive) can be obtained by plotting more
than cumulative frequencies against lower limits of the class intervals. As we
have already mentioned within brackets that less than cumulative frequency
curve and more than cumulative frequency curve are also called less than
ogive and more than ogive respectively.
In other words we may define less than ogive and more than ogive as follow:
Less Than Ogive: If we plot the points with the upper limits of the classes as
abscissae and the cumulative frequencies corresponding to the values less then
the upper limits as ordinates and join the points so plotted by line segments, the
curve thus obtained is nothing but known as “less than cumulative frequency
curve” or “less than ogive”. It is a rising curve.

54
More Than Ogive: If we plot the points with the lower limits of the classes as Graphical Presentation of
abscissae and the cumulative frequencies corresponding to the values more Data-I
than the lower limits as ordinates and join the points so plotted by line
segments, the curve thus obtained is nothing but known as “more than
cumulative frequency curve” or “more than ogive”. It is a falling curve.
Let us draw both the ogives (‘less than’ and ‘more than’) for the following
frequency distribution of the weekly wages of number of workers given in
Table 15.5.

Table 15.5
Weekly 0-10 10-20 20-30 30-40 40-50
wages
No. of 45 55 70 40 10
workers

Before drawing the ogives, we make a cumulative frequency distribution as

given in table 15.6

Table 15.6
Weekly No. of Less than Cumulative More than Cumulative
wages workers frequency distribution frequency distribution
Wages Number of Wages Number of
Less than workers More than workers
0-10 45 10 45 0 220
10-20 55 20 100 10 175
20-30 70 30 170 20 120
30-40 40 40 210 30 50
40-50 10 50 220 40 10

From above data, we construct both the ogives as shown in Fig. 15.8 and Fig.
15.9:

Fig. 15.8: Less Than Ogive.

55
Presentation of Data

Fig. 15.9: More Than Ogive.

For “less than ogive” as shown on previous page in Fig. 15.8, we have plotted
the points (10, 45), (20,100), (30, 170), (40, 210), (50, 220) and then joined
them by line segments. Similarly, for “more than ogive” as shown above in
Fig. 15.9, we have plotted the points (0, 220), (10, 175), (20, 120), (30, 50),
(40, 10), and then joined them by line segments.
If we want to obtain a partition value, using ogives, we draw dotted horizontal
line through that value at y-axis which corresponds to the partition value and
then from the point, where it meets the less then ogive, we draw a dotted
vertical line and let it meets the x-axis. The abscissa of the point, where it
meets the x-axis is the required partition value. For example, suppose we want
to find first quartile, then we draw a dotted horizontal line starting from y-axis
at a point corresponding to N/4 and let it meets the “less than ogive”. From that
point at “less than ogive”, we draw a dotted vertical line and let it meets the x-
axis. The abscissa corresponding to this point is the first quartile. Similarly, for
finding median or second quartile, we start drawing dotted horizontal line from
y-axis at a point corresponding to N/2 and then we proceed as described above.
Similarly, for third quartile 3N/4 is taken in place of N/2. In this way, we may
find any partition value.
Note 4: Median may also the obtained by drawing dotted vertical line through
the point of inter section of both the ogives, when drawn on a single figure.
Now, you can try the following exercises.
E5) Draw two ogives from the following data
Class: 0-10 10-20 20-30 30-40 40-50 50-60 60-70 70-80 80-90
Frequency: 3 6 10 13 20 18 15 9 6
Hence find median. Compare your result by calculating median by direct
calculatios.
E6) Draw less than ogive from the following frequency distribution of marks
of 90 students
Marks: 0-9 10-19 20-29 30-39 40-49 50-59 60-69 70-79
No. of Students: 7 11 19 8 20 14 8 3
Hence find Q1, Q2 and Q3.
E7) Draw the more than ogive for the following frequency distribution of the
weekly wages of workers:
Weekly wages: 0-10 10-20 20-40 40-50 50-60 60-70 70-80 80-90 90-100
No. of Workers: 5 15 20 30 45 35 25 15 10

56
15.4 SUMMARY Graphical Presentation of
Data-I
In this unit we have discussed:
1) Various types of graphical presentation of data.
2) Way of drawing histogram for continuous frequency distributions.
3) Frequency polygon for a frequency distribution.
4) Frequency curves of different shapes, and
5) Way of drawing cumulative frequency graphs or ogives.

15.5 SOLUTIONS/ANSWERS

E1) Histogram of the given data is given below:

E2) Histogram of the given data is given below:

E3) Frequency polygon of the given data by first drawing histogram is given
on the next page.

57
Presentation of Data

Frequency polygon can also be drawn without histogram as given below:

Note 5: We can draw frequency polygon by first drawing a histogram or

we can draw directly.

E4) Frequency curve for the given data is given below.

30

25

20

15
Y-Values
10

0
0 2 4 6 8 10

E5) Two ogives for the given data are given on the next page.

58
 Graphical Presentation of
Data-I

Calculations of Median through direct calculation:

Given data are

Class Interval Frequency (f) Cumulative frequency (cf)

(CI)
0-10 3 3
10-20 6 9
20-30 10 19
30-40 13 32
40-50 20 52
50-60 18 70
60-70 15 85
70-80 9 94
80-90 6 100
Total 100

Here N   f  sum of all frequencies 100

N 100
  = 50 and hence median class is 40-50.
2 2
So, in usual notations we have
l = lower limit of the median class = 40,
h = width of the median class = 50  40 = 10
c = cumulative frequency preceding the median class = 32
f = frequency of the median class = 20
hN  10  100  1
Now, Median  l    c   40    32   40   50  32 
f2  20  2  2
1
 40  18   40  9  49
2
Thus we see that median obtained by using two ogives graphically and by
direct calculation are same.
59
Presentation of Data
E6) Less than ogive for the given frequency distribution is given below.
N 90
To find Q1 :   22.5 first of all we have to draw a dotted
4 4
 N 
horizontal line starting from y-axis at 22.5  as  22.5  and then from
 4 
the point where it meets the less ogive, we shall draw dotted vertical line,
the value corresponding to the point where it meets the horizontal axis is
the value of Q1 as shown in figure. Similarly, values of Q 2 and Q 3 can
be obtained as shown in the figure below:

E7) More than ogive for the given frequency distribution is given below:

60