MEASURES OF DISPERSION

Structure:
1.0 Introduction
1.1 Objectives
1.2 Requisites of Good Measures of Dispersion
1.3 Methods of Studying of Dispersion
1.3.1 Range
1.3.2 Quartile Deviation
1.3.3 Mean Deviation
1.3.4 Standard Deviation

1.4 Exercise

1.0 INTRODUCTION
We know how to sum up the data into a single representative value. However, that
value does not reveal the variability present in the data. In this module we will study
those measures, which seek to quantify variability of the data. It is quite obvious
that averages try to tell only one aspect of a distribution i.e. a representative size of
the values. To understand it better, you need to know the spread of values also.
Dispersion: is the extent to which values in a distribution differ from the average of
the distribution.

1.1 OBJECTIVES
Measures of dispersion serve the following objects.
● To determine the reliability of an average
● To compare the variability of different distributions
● To control the variability

1.2 REQUISITES OF GOOD MEASURES OF DISPERSION

A good measure of dispersion should have the following properties
● It should be simple to understand and rigidly defined
● It should be easy to compute
● It should be based on all values
● It should be capable of further algebraic treatment
● It should have sampling stability
● It should not be unduly affected by extreme items

1.3 METHODS OF STUDING DISPERSION

2 | Page
1.3.1 RANGE
Range (R) is the difference between the largest (L) and the smallest value (S)
in a distribution. Thus, R = L – S
Higher value of Range implies higher dispersion and vice-versa.
The corresponding relative measure, called the coefficient of range, is given
by

Coefficient of range , where L and S have the above meaning

Example 1: Look at the following values and calculate the Range.

20, 30, 40, 50, 200
Solution : Here, L=200, S=20 and R=L-S=200-20=180,
Therefore Range (R) = 180
• What is the Range if the value 200 is not present in the data set?
If 200 is not in data set, New data set is 20, 30, 40, 50.
Here, L=50,S=20 and R=50-20=30, Therefore Range (R) = 30
• If 50 is replaced by 150, what will be the Range?
New data set if 50 is replaced by 150 is, 20, 30, 40, 150, 200
Here, L=200, S=20 and R=L-S=200-20=180, Therefore Range (R) = 180
Merits :
● Range is easy to understand
● Easy to calculate
Demerits :
● No sampling stability
● Does not depend upon all the values
Computation of range:
(a) Discrete Distribution
Example 2: The following are the marks obtained by six students in
statistics. Calculate the range and coefficient of range.
Sr. No. 1 2 3 4 5 6
Marks 30 35 40 80 70 62
Solution: Range = L-S, L=80, S=30
Range = 80-30 =50 Marks

Coefficient of range =
(b) Continuous distribution
In continuous distribution , the range is the difference between the mid-point
of the highest class and that of the lowest class.
Example 3: Find the range and coefficient of range from the following data

Weight in lbs 80-90 90-100 100-110 110-120 120-130

No. of persons 4 8 12 14 7

3 | Page
Solution: Range= L-S, but L= 125, the mid-point of highest class and
S=85, the mid-point of the lowest class. Range= 125-85=40 lbs

Coefficient of range =

1.3.2 QUARTILE DEVIATION

The presence of even one extremely high or low value in a distribution can
reduce the utility of range as a measure of dispersion. Thus, you may need a
measure which is not unduly affected by the outliers. In such a situation, if
the entire data is divided into four equal parts, each containing 25% of the
values, we get the values of Quartiles and Median. The upper and lower
quartiles (Q3 and Q1, respectively) are used to calculate Inter Quartile Range
which is Q3 – Q1. Inter-Quartile Range is based upon middle 50% of the
values in a distribution and is, therefore, not affected by extreme values. Half
of the Inter-Quartile Range is called Quartile Deviation (Q.D.). Thus:

Q.D. is therefore also called Semi-Inter Quartile Range. Q.D. is absolute

measure of dispersion. The corresponding relative measure, called the
coefficient of Quartile Deviation, is given by

Merits :
● Simple to understand
● Easy to calculate
● It is not affected by extreme values
● It is especially useful to measure the variation of a distribution with
open end classes.
Demerits :
● It is not based on all observations
● It is not capable of further algebraic treatment
● It is affected by sampling fluctuations
● It ignores the first 25% and the last 25% items

Computation of Quartile Deviation: (ungrouped data)

Example 4: Calculate Range and Q.D. of the following observations:

20, 25, 29, 30, 35, 39, 41, 48, 51, 60 and 70
Solution: For Q.D., we need to calculate values of Q3 and Q1.

4 | Page
Here n = 11

Q1 (25 th percentile) =
= Size of (11+1)/4
= size of the 3rd item
So, Q1 = 29

Q3 (75 percentile) =
= Size of [3(11+1)]/4
= size of the 9th item
So, Q3 = 51

Computation of Quartile Deviation: (grouped data)

Examples 5: For the following distribution of marks scored by a class of 40
students , Compute the quartile deviation and coefficient of Q.D.

Class : 0-10 10-20 20-30 30-40 40-50

Frequency : 5 8 16 7 4

Solution:
Class Frequency Cumulative
Frequency
0-10 5 5
10-20 8 13
20-30 16 29
30-40 7 36
40-50 4 40

40

5 | Page
Therefore Quartile class is 10-20
Where, =10, =5, = 8, =10 and =40

Therefore Quartile class is 30-40

Where, =30, =29, = 7, =10 and =40

6 | Page
 1.3.3 MEAN DEVIATION

Range and Quartile Deviation are not useful in measuring, how far the values are,
from their average. Yet, by calculating the spread of values, they do give a good
idea about the dispersion. Two measures which are based upon deviation of the
values from their average are Mean Deviation and Standard Deviation. Since the
average is a central value, some deviations are positive and some are negative. If
these are added as they are, the sum will not reveal anything. In fact, the sum of
deviations from Arithmetic Mean is always zero. Mean deviation depends upon all
the figures and therefore, is superior to them.

Definition: Mean deviation is defined as the arithmetic mean of the absolute

deviation of all the values from any central value. Generally the deviations are taken
from median. This is advantageous because the sum of the deviation is the least if
they are taken from the median.

Symbolically, Mean Deviation

Where, = the deviation from median

=total number of items
The corresponding relative measure, called the coefficient of mean deviation, is
given by

Coefficient of Mean Deviation

Minimal Property of Mean Deviation :
The mean deviation is minimum when deviations are taken from median. The

property says that is minimum, when is median.

Merits :
● It is easy to understand
● Easy to calculate
● It is based on all the observations
● It is less affected by the end-values
Demerits :
● It is not capable of further algebraic treatment

7 | Page
 ● It does not give reliable results in many cases, particularly when the
degree of variability is very high.

Computation of Quartile Deviation: (ungrouped data)

Example 6: Calculate the Mean Deviation from median and its coefficient from the
following data
Salaries in Rs. 150, 210, 240, 280, 300, 330, 360
Solution:
Steps:
(i) The median of the values is calculated
(ii) Difference between each value and the median is calculated. All differences are
considered positive. These are denoted as |d|
(iii)The A.M. of these differences (called deviations) is the Mean Deviation.

Median is =280

Salaries

150 -130 130

210 -70 70
240 -40 40
280 0 0
300 20 20
330 50 50
360 80 80
n=7
=390

Mean Deviation

Coefficient of Mean Deviation

Computation of Quartile Deviation: (grouped data)

8 | Page
Example 7: Calculate the Mean Deviation from median and its coefficient from the
following data

Marks : 13 14 15 16 17
No. of students : 4 8 12 9 6
Solution:

Marks

Class Frequency Cumulative

Frequency
13 4 4 2 8
14 8 12 1 8
15 12 24 0 0
16 9 33 1 9
17 6 39 2 12
=39 =37

Mean Deviation

Coefficient of Mean Deviation

1.3.4 STANDARD DEVIATION

Standard Deviation is the positive square root of the mean of squared deviations
from mean. So if there are five values x1, x2, x3, x4 and x5, first their mean is
calculated. Then deviations of the values from mean are calculated. These
deviations are then squared. The mean of these squared deviations is the variance.

9 | Page
Positive square root of the variance is the standard deviation. (Note that Standard
Deviation is calculated on the basis of the mean only).

Definition: Standard deviation is defined as ‘the square-root of the arithmetic mean

of the squares of the deviations from the mean’. It can be therefore, defined as the
root-mean square deviation about mean. It is denoted by . Symbolically,

Where, , the deviations of from and

If the deviations are taken by from any value A, the square root of the mean of the
squares of the deviations from is called root-mean square deviation and is
denoted by .

Hence, and is minimum if

The square of is called mean square deviation, thus

Merits :
● It is rigidly defined
● It is based on all the observations
● It is less affected by sampling fluctuations
● It can be treated algebraically in further analysis
● It is more accurate hence more reliable
Demerits :
● Not easy to compute
● It gives more weights to extreme items and small weights to items
nearer to it
Variance :
Square of standard deviation is called variance. Variance has its own place in
statistical analysis. It is denoted by

10 | Page
Coefficient of variation (C.V.)

The relative measure of standard deviation is coefficient of standard deviation and is

given by

Better way to compare the variability of different groups is to express this ratio in
percentage. This is called “coefficient of variation”.

Symbolically,
If we want to compare the variability of different groups, we compare their
coefficient of variation. The higher the coefficient of variation, the higher is the
variability.

Absolute and relative measures of dispersion

Range, Quartile deviation, Mean deviation and Standard deviation are absolute
measures of dispersion, because they are expressed in the units in which the data are
expressed. They are useful to measure the variability of a given group. To compare
the variability of different groups absolute measures of dispersion are not useful. We
calculate the ratio of the measure of dispersion to some other convenient number.
This ratio is called ‘the relative measure of dispersion’ and is a pure number free
from units. They are also called relative measures of dispersion. If the relative
measure of dispersion of one group is higher than the other, then the group of higher
relative measure of dispersion has higher variability. All the absolute and relative
measures of dispersion are given below.

Absolute Measures Relative Measures

1 1

2 2

3 3

4 4

11 | Page
 5

Derivation of different formulae for standard deviation

(a) To find from actual values

Since,

(b) To find by taking deviation from an assumed mean

where,

(c) Effect of change of origin and change of scale on S.D.

Let

Thus, the standard deviation is not affected by change of origin but is affected by
change of scale.

(d) To find the S.D. by change of origin and change of scale.

12 | Page
If we write,

Property: If two distributions have means and , standard deviations ,

and sizes , , then the standard deviation of the combined distribution is
given by

Where, Combined mean , ,

Computation of Standard deviation

(a) Individual items

The S.D. can be calculated by taking deviations either from actual mean or from an
assumed mean by using the formula.

Example 8. : Calculate the standard deviation of the following values:

5, 10, 25, 30, 50

Solution: Mean =

5 -19 361
10 -14 196
25 1 1

13 | Page
 30 6 36
50 26 676
0 1270

(b) Discrete distribution

The S.D. can be calculated by taking deviation from assumed mean. If
Where, is assumed mean. Then the formula to find S.D. is,

Example 9. : Calculate the standard deviation of the following values:

x : 7 8 9 10 11 12 13
f : 4 6 9 12 9 6 4
Solution: Here, take =9 and

7 4 -2 -8 4 16
8 6 -1 -6 1 12
9 9 0 0 0 0
10 12 1 12 1 12
11 9 2 18 4 36
12 6 3 18 9 54
13 4 4 16 16 64

Now,

Hence,

14 | Page
(c) Continuous distribution
In continuous distribution change of origin and scale method is used to compute

S.D. Here, , is assumed mean and is scale

Example 10. : Calculate the standard deviation of the following values:

x : 0-10 10-20 20-30 30-40 40-50
f : 5 8 15 16 6
Solution: Here, take =25, and

0-10 5 5 -2 -10 4 20
10-20 15 8 -1 -8 1 8
20-30 25 15 0 0 0 0
30-40 35 16 1 16 1 16
40-50 45 6 2 12 4 24
68
10

1.4 Exercise

Q1.: The yield of wheat per acre for 10 districts of a state is as under:
District : 1 2 3 4 5 6 7 8 9 10
Wheat : 12 10 15 19 21 16 18 9 25 10
Calculate (i) Range
(ii) Q.D.
15 | Page
 (iii) Mean Deviation about Mean
(iv) Mean Deviation about Median
(v) Standard deviation

Q2.: A batsman is to be selected for a cricket team. The choice is between X and Y
on the basis of their scores in five previous tests which are:
X : 25 85 40 80 120
Y: 50 70 65 45 80
Which batsman should be selected if we want,
(i) a higher run getter, or
(ii) a more reliable batsman in the team?

Q3.: Calculate the Mean Deviation using mean and Standard Deviation for the
following distribution.
Classes : 20-40 40-80 80-100 100-120 120-140
Frequencies : 3 6 20 12 9

Q4.: The sum of 10 values is 100 and the sum of their squares is 1090. Find out the
Coefficient of Variation.

------------------------------------------------MKP-----------------------------------------------

16 | Page