ABM 401 Lesson 9

Lesson 9
Measures of Dispersion I:
Range and Quartile Deviation

Objectives of the Lesson:

To make students to understand:
The concept and properties of dispersion
Objectives of measuring Dispersion
Measures of dispersion: Range and Quartile Deviation

INTRODUCTION
Generally idea of central part of the distribution can be known through averages or
measurement of central tendency. If we arrived to average, it can not give complete
idea about the distribution as will be clear from the following example:
Consider the series: (i) 7,8,9,10,11 (ii) 3,6,9,12,15 (iii) 3,5,9,13,17

In all above cases we see that the number of observation is 5 and the mean is 9. If
we are given that mean of 5 observations is 9, we can not form an idea as to
whether it is average of first series or second series or third series or of any other
series of 5 observations whose sum is 45. Thus we see that the measures of
central tendency are inadequate to give us complete idea of the distribution. They
must be supported and supplemented by other measures. One such measure is
dispersion.

Dispersion means “scatteredness”. We study dispersion to have an idea about

homogeneity of the distribution. In the above example we say that series i) is more
homogeneous (less dispersed) than the series (ii) or (iii) or we say that series (iii) is
more heterogeneous (more scattered) than the series (i) or (ii).

DISPERSION: MEANING, DEFINITION AND PROPERTIES

One of the greatest defects of the averages of the first order is that they fail to
represent all the peculiarities and characteristics of the series itself. There are
other anomalies also which ask for a better measure. In general, to achieve this
123
ABM 401 Lesson 9

objective of having a better and more appropriate result the averages of the second
order are calculated. They remove the drawbacks of the averages of the first order
and yield a better and more representative result.

An average of the second order is an average of the differences of all the items of
the series from an average of those items. In averaging these differences their
irregularities are brushed off and a representative figure results. Dispersion is a
term which is very commonly used in the study of statistics. Its dictionary meaning
includes 'fluctuation', 'scatter', 'variation', 'variability', 'diversity', 'deviation' or
'spread'. The common usage, however, is 'scatter' and 'deviation'.

The simplest meaning given to the word 'dispersion' is a lack of uniformity in the
sizes or qualities of the items of a group or series. The word dispers ion may also
be used to indicate the spread of the, data. Normally under dispersion, we
concentrate on the approach of measuring the extent and nature of scatteredness
and not on the direction of scatteredness. Seigleman has definition dispersion as:
"Dispersion is the extent to which the magnitudes or qualities of the items differ:
that is, the degree of diversity." Thus, "The property which denotes the extent to
which the values are dispersed about the central value is termed as dispersion."

Use of Measures of Dispersion:

Average is a typical value but it alone does not describe the data fully. It does not
tell us how the items in a series are scattered around it. For example, in two cities
the mean family income may be the same, say, Rs. 10,000 a year but in one we
find a middle class population with a large majority of annual income approximately
Rs. 10,000, whereas in the other city we find a wide inequality in income, with very
poor and very rich population. Thus, the mean income would not reveal the wide
difference in their social and economic structure,

Therefore, in order to know how representative the average is of statistical data we

should also know the extent over which the other values are spread around it.
Thus, a measure of dispersion helps us in gauging the representativeness of the
average. It is also used for comparing the variability or consistency (uniformity) of
124
ABM 401 Lesson 9

two or more series. A higher degree of variation means smaller consistency. It is

also used in order to estimate the accuracy of a test result or to test its
significance. In short, along with an average, it describes the structures of the
distribution.

Properties of an Ideal Measure of Dispersion:

The key elements for an ideal measure of dispersion are same as those for an
ideal measure of central tendency viz.:
It should be readily comprehensible;
Be able to utilize all the data;
Possible to calculate with reasonable ease and rapidity;
Afford a good standard of comparison;
Be suitable for algebraic and arithmetical manipulation; and
It should be very little affected by variations of sampling. .

Absolute and Relative Dispersion:

Like any other mathematical concept, dispersion, too, can be ascer tained in either
the absolute or in the relative form. It is, however, to be remembered that
dispersion, in its true sense, is studied in its absolute form. None can challenge the
validity of the use of the study of dispersion in its absolute form. It is on account of
the fact that even if the measure is absolute, it can provide the requisite result.
Some of the measures of dispersion, viz., the range and the mean deviation are
theoretically incorrect and in practice give non-uniform and un-reliable results. On
account of it, they are criticized and some go as far as to mention the concept of
dispersion as erroneous and imaginary. But, contrary to it, there is the standard
deviation and the variance that can very easily provide a better and more reliable
result.

Measures of Dispersion:
Averages of the second order or the dispersion of a series may be calculated by
the application of any of the following methods:

125
ABM 401 Lesson 9

1. Range and Coefficient of Range;

2. Inter-Quartile Range and Quartile Deviation;
3. Coefficient of Quartile Deviation;
4. Mean Deviation or the average deviation; Consisting of:
Mean deviation from the arithmetic mean;
Mean deviation from the mode; and
Mean deviation from the median;
5. Standard Deviation;
6. Lorenz curve (The Graphic Method).

In this lesson only first three measures are covered.

RANGE
Of all the measures of dispersion, range is the simplest. In this case, the maximum
and the minimum values (or the greatest and the least values) are selected and
finally their differences are known. It gives the value of the Range. If it is being
calculated for a continuous frequency distribution, it is the difference between the
lower limit of the smallest class and the upper limit of the biggest class.
Symbolically: R = (L – S)
Where: R is Range, L is the largest value and S is the smallest value.

Coefficient of Range: If the range so calculated is divided by the sum of the two
extreme limits, the result is known as the coefficient of the scatter, or Coefficient of
Range (CR).

Merits of Range: The range method of finding the value of dispersion and its
coefficient has some advantages. For this, in spite of its certain inherent defects,
the method is put to some use. Some of its merits are:
Simplest measure of dispersion;
Easy to calculate;
Easy to understand clearly.
126
ABM 401 Lesson 9

Demerits of Range: Most of the statisticians hesitate to recognize the method of

Range as a measure of dispersion. Their doubt to some extent is correct. Some of
the drawbacks of this system of measurement may be mentioned as:
The Range takes no account of the form of distribution within the range.
As only two values, the greatest and the least are taken into consid eration, it
is affected by extreme values.
The range is subject to fluctuations if more than one sample is taken into
consideration.
This measure is termed as inadequate because It fails to consider all the
item values.
The result obtained from the Range method cannot be used for fur ther
statistical treatment.
Mathematically, this method is defective, because all the items of the
variable are not taken into consideration.

Use of Range:
It is commonly used particularly in those fields where variation is not much,
e.g., in quality control of manufactured items.
It is fruitfully used in studying the variations in money rates and rates of
exchange.

The Inter Quartile Range is a measure of first order dispersion. This method may
also be used to find the coefficient or the relative measure. The values of the
variable which divide the distribution in the ratio of 1:3 and 3:1 are called the
quartiles. Symbolically, the Inter-quartile Range is:

IQR = (Q 3 - Q 1)
The quartiles and the median divide the distribution into four equal parts. Half of
the difference between the first and the third quartiles is called the quartile
deviations.

The quartile deviation, as the name suggests, is dependent on the values of Q 1 and
Q3 In general, this method of dispersion may easily be put to use if very accurate
127
ABM 401 Lesson 9

results are not required. The quartile deviation, however, cannot take into
consideration all the items. Thus, it excludes items of both the extremes and takes
into consideration only the central values. Symbolically, the Quartile Deviation is:

Or

Where, M is Median, Q1 and Q 3 are First and Third Quartile respectively.

Merits of Quartile Deviation: Some of the advantages of the measure of

dispersion (or semi-interquartile range) may be summed up as:

It is easy to calculate. We are required simply to find the values of Q1 and Q3

and then to apply the formula. The coefficient too can be easily calculated.
It gives a better result than the range method as it makes use of 50% of the
data.
The quartile deviation is not affected by the extreme values.

Disadvantages or the Quartile Deviation: In spite of these merits of this system,

it is generally not used for making comparisons or to find the values of the
dispersion and its coefficient. The reasons are:

It considers only 50% data and ignores other 50% variables. It is mainly
dependent on the values of the central items. If these values are irregular
and abnormal, the result is bound to be affected.
All the items of the frequency distribution cannot be given equal weight or
importance in finding the values of Q1 and Q3 or the quartile deviation.
It is an inaccurate measure of dispersion. It can not be regarded as reliable
measures. Hence, it should not be used unless asked for.

128
ABM 401 Lesson 9

Example 1: Annual incomes of three groups of families are given below. Calculate
Range and Relative Range.
Family Group I Family Group II Family Group III
16800 15200 13000
16800 16000 25000
16800 16800 25000
16800 17600 14000
16800 18400 7000

Solution: In the above three groups, the range is:

I series: R = (Rs. 16800 - Rs. 16800) = 0
II series: R = (Rs. 18400 – Rs. 15200 = 3200
III series: R = (Rs. 25000 – Rs. 7000 = 18000
The relative measure of dispersion corresponding to range is arrived at as follows:

Group I

Group II

Group III

Example 2: Calculate quartile deviation and its coefficient of A’s monthly earnings
for a year.
Month 1 2 3 4 5 6 7 8 9 10 11 12
Earning 239 250 251 251 257 258 260 261 262 262 273 275

3.25th month means Rs. 251. So, Q1 = 251

129
ABM 401 Lesson 9

9.75th month means Rs. 262. So, Q3 = 262

Example 3: Calculate the Semi inter-quartile range and quartile coefficient from the
following.

Age in Years 20 30 40 50 60 70 80
No. of members 3 61 132 153 140 51 3
Solution:

Age in Years (X) 20 30 40 50 60 70 80

No. of members (F) 3 61 132 153 140 51 3
CF 3 64 196 349 489 540 543

136th item is in CF 196 which has corresponding item value 40. So, Q1 =40 Years

408th item is in CF 489 which has corresponding item value 60. So, Q3 =60 Years

Example 4: Calculate the range and semi-inter quartile range of wages. Also
calculate the quartile coefficient of dispersion.
Wages (Rs.) 30-32 32-34 34-36 36-38 38-40 40-42 42-44
No. of members 12 18 16 14 12 8 6
Solution:
Wages (Rs.) 30-32 32-34 34-36 36-38 38-40 40-42 42-44
Members (F) 12 18 16 14 12 8 6
(CF) 12 30 46 60 72 80 86

130
ABM 401 Lesson 9

21.5th item is in 32.-34 class, so value of by formula:

64.5th item is in 38-40 class, so value of by formula:

SUMMARY

Averages of the second order are called as dispersion of a series. There are
various measures of dispersion like Range, Coefficient of Range, Inter-Quartile
Range, Quartile Deviation and its Coefficient, Mean Deviation, Standard Deviation,
and also the Lorenz curve. Range is the difference between the lower limit of the
smallest class. Inter Quartile Range is a measure of first order dispersion. The
values of the variable which divide the distribution in the ratio of 1:3 and 3:1 are
called the quartiles. We know that quartiles and the median divide the distribution
into four equal parts. Half of the difference between the first and the third quartiles
is called the quartile deviations.

REVIEW QUESTIONS

1. What do you mean by dispersion? Explain clearly methods of measuring

dispersion and its uses.

131
ABM 401 Lesson 9

2. Write short notes on: (i) Range; (ii) Quartile Deviation; (iii) Coefficient of Quartile
Deviation.
3. Find Quartile deviation and its coefficient from the following data.
Age 15 16 17 18 19 20 21
No. of students 4 6 10 15 12 9 4
Ans. Q1=17, Q3=19, QD=1, Co. QD=0.056
4. Calculate Inter-quartile Range, Quartile Deviation, and Quartile Coefficient of
Dispersion from the following data
Marks 20 30 40 50 60 70 80
No. of students 3 61 132 153 140 51 3
Ans. Q1=40, Q3=60, IQR=20, QD=10, Co. QD=0.2
5. Calculate Quartile Deviation and Coefficient of Quartile Deviation from the
following data
Height (Inches) 50-53 53-56 56-59 59-62 62-65 65-68
No. of students 2 7 24 27 13 3
Ans. Q1=57.25, Q3=61.67, QD=2.21, Co. QD=0.037

SUGGESTED READINGS

Statistical Methods: SP Gupta

Fundamentals of Statistics: DN Elhance
Statistics: BN Gupta
Statistics: RP Hooda
Fundamentals of Statistics: KN Nagar

132