Measures of Dispersion

MEASURES OF DISPERSION

The previous lesson provides the measure of central tendency that sum up or
describe the data into a single representative value. The measures of
central tendency may not be adequate to describe data unless we know the
manner in which the individual items scatter around it. In other words, a further
description of a series on the scatter or variability known as dispersion is
necessary, if we are to gauge how representative the average is.
Let us take the following three sets.
Students Group X Group Y Group Z
1 50 45 05
2 50 50 45
3 50 55 100
mean X 50 50 50
Thus, the three groups have same mean i.e. 50. In fact the median of group X and
Y are also equal. Now if one would say that the students from the three groups are
of equal capabilities, it is totally a wrong conclusion. Close examination
reveals that in group X students have equal marks as the mean, students from
group Y are very close to the mean but in the third group Z, the marks are widely
scattered. It is thus clear that the measures of the central tendency is alone not
sufficient to describe the data. The measure of dispersion helps us to know
the degree of variability in the data and provide a better understanding of the
data.

OBJECTIVES
After completing this lesson, you will be able to:
 know the meaning and need of measures of dispersion;
Measures of Dispersion

 distinguish between absolute and the relative measures of dispersion;

 apply the various measures of dispersion; and
 calculate and compare the different measures of dispersion.

9.1 MEANING OF DISPERSION

Dispersion is the extent to which values in a distribution differ from the average
of the distribution.

In measuring dispersion, it is imperative to know the amount of variation (absolute

measure) and the degree of variation (relative measure). In the former case we
consider the range, Quartile Deviation, standard deviation etc. In the latter case
we consider the coefficient of range, coefficient quartile deviation, the coefficient
of variation etc.

9.1.1 Absolute and Relative Measures of Dispersion

The dispersion of a series may be measured either absolutely or relatively. If the
dispersion is expressed in terms of the original units of the series, it is called
absolute measure of dispersion. The disadvantage of absolute measure of dispersion
is that it is not suitable for comparative study of the characteristics of two or more
series.

For example if the data is expressed in kilograms then the absolute variation will
also be expressed in kilograms but if the same data is expressed in grams then the
variation will appear 1000 times more. So for comparison point of view it is
necessary to calculate the relative measures of dispersion which are expressed as
percentage form (i.e. unitless number). These types of expressions are called
coefficients of dispersion. Each absolute measure of dispersion has a relative
measure of dispersion.

9.2 MEASURES AND METHODS OF COMPUTING

DISPERSION
The following are the important measures of dispersion:
1. Range
2. Quartile deviation or Semi-Inter quartile range.
3. Mean deviation
4. Standard deviation
5. Lorenz Curve
 Measures of Dispersion

Range and Quartile Deviation measure the dispersion by calculating the spread
within which the values lie. Mean Deviation and Standard Deviation calculate the
extent to which the values differ from the average.

9.2.1 Range
Range (R) is the difference between the largest (L) and the smallest value (S) in
a distribution. Thus

Range (R) = L – S
Coefficient of Range: It is a relative measure of the range. It is used in the
comparative study of the dispersion

L−S
co-efficient of Range =
L+S

In case of continuous series Range is just the difference between the upper limit
of the highest class and the lower limit of the lowest class.

Range: Evaluation
Range is very simple to understand and easy to calculate. However, it is not based
on all the observations of the distribution and is unduly affected by the extreme
values. Any change in the data not related to minimum and maximum values will
not affect range. It cannot be calculated for open-ended frequency distribution.

Example 1: The amount spent (in `) by the group of 10 students in the school
canteen is as follows:

110, 117, 129, 197, 190, 100, 100, 178, 255, 790.
Find the range and the co-efficient of the range.
Solution: R = L - S = 790 - 100 = ` 690

L −S 790 −100 690

Co-efficient of Range = = = = 0.78
L+S 790 +100 890

Example 2: Find the range and it’s co-efficient from the following data.
Size 10-20 20-30 30-40 40-50 50-100
Frequency 2 3 5 4 2

Solution: R = L – S = 100 – 10 = 90
Measures of Dispersion

L −S 100 −10 90
Co-efficient of range = = = = 0.82
L+S 100 +10 110

INTEXT QUESTIONS 9.1

1. The difference between the largest and the smallest data values is the
(a) variance (b) inter-quartile range
(c) range (d) coefficient of variation
2. Aresearcher has collected the following sample data. The mean of the sample
is 5.
3 5 12 3 2
The range is
(a) 1 (b) 2 (c) 10 (d) 12

9.2.2 Quartile Deviation

It is based on the lower quartile Q1 and the upper quartile Q3. The difference
Q3 – Q1 is called the inter-quartile range. The difference Q3 – Q1 divided by 2 is
called semi-inter-quartile range or the quartile deviation.

Q3 − Q1
Thus Quartile Deviation (Q.D) =
2

9.2.2.1 Coefficient of Quartile Deviation

A relative measure of dispersion based on the quartile deviation is called the
coefficient of quartile deviation. It is defined as

Q3 − Q1
Q − Q1
Coefficient of Quartile Deviation = 2 = 3
Q3 + Q1 Q3 + Q1
2

It is pure number free of any units of measurement. It can be used for comparing
the dispersion in two or more than two sets of data.

9.2.2.2 Computation of Quartile Deviation of Ungrouped Data

Example 3: Find out the quartile deviation of daily wages (in `) of 7 persons is
given below:120,70,150,100,190,170,250
 Measures of Dispersion

Solution:
Arranging the data in an ascending order we get
70, 100, 120, 150, 170, 190, 250

Here n = 7,

(N +1)
Q1 = Size of th item
4

(7 +1)
= Size of th item = 2nd item = 100 rupees
4

3(N +1)
Q3 = Size of th item
4

3(7 +1)
= Size of th item = 6th item = 190 rupees
4
Q3 − Q1 190 −100
Q.D. = = = 45 rupees
2 2

INTEXT QUESTION 9.2

1. If the first quartile is 104 and quartile deviation is 18. Find the third quartile.

Example 4: The wheat production (in Kg) of 20 acres is given as: 1120, 1240,
1320, 1040, 1080, 1200, 1440, 1360, 1680, 1730, 1785, 1342, 1960, 1880, 1755,
1720, 1600, 1470, 1750, and 1885. Find the quartile deviation and coefficient of
quartile deviation.

Solution:
After arranging the observations in ascending order, we get
1040, 1080, 1120, 1200, 1240, 1320, 1342, 1360, 1440, 1470, 1600, 1680, 1720,
1730, 1750, 1755, 1785, 1880, 1885, 1960.

( N +1 
Q1 = value of   th item
4 )

( 20 +1 
= Value of   th item
4 )
Measures of Dispersion

= Value of (5.25)th item

= 5th item + 0.25(6th item – 5th item)
= 1240 + 0.25(1320 – 1240)
Q1 = 1240 + 20 = 1260
Q1 = 1240 + 20 = 1260 kg

3( N + 1)
Q3 = Value of th item
4

3(20 +1)
= Value of th item
4

= Value of (15.75)th item

= 15th item + 0.75(16th item – 15th item)
= 1750 + 0.75 (1755 – 1750)
Q3 = 1750+3.75 = 1753.75 kg
Quartile Deviation (Q.D)

Q3 − Q1 = 1753.75 −1260 = 492.75 = 246.875

=
2 2 2

Q3 − Q1 = 1753.75 −1260
Q.D. =
2 2

492.75
= = 246.875kg.
2

Coefficient of Quartile Deviation

Q3 − Q1 1753.75 −1260
= = = 0.164
Q3 + Q1 1753.75 +1260

Computation of Q.D. for a frequency distribution

9.2.2.2 Computation in case of Discrete Series:

Example 5: The Tax authority collected the following amount of tax from
different firms in a particular market.
 Measures of Dispersion

Amount of Taxes (in 000 `) 10 11 12 13 14

No. of Firms 3 12 18 12 3
Calculate the quartile deviation and the coefficient of quartile deviation.

Solution:
Table 9.1: Calculation of Quartile deviation
Amount of Taxes No. of Firms (f) Cummulative
(in ’000 `) Frequency (C.F.)

10 3 3
11 12 15
12 18 33
13 12 45
14 3 48
f = 48

Here N = 48,

(N + 1) th item
Q1 = Size of
4

(48 + 1) th item
= Size of
4

= Size of12.25th item = 11 (in '000 rupees)

3(N + 1)
Q3 = Size of th item
4

3(48 + 1)
= Size of th item
4

= Size of 36.75th item = 13(in '000 rupees)

Q3 − Q1 13 −11
Q.D. = = = 1 (in '000 rupees)
2 2
Measures of Dispersion

Q3 − Q1 13 −11
Coeff of Q.D. = Q + Q = 13 +11 = 0.083
3 1

9.2.2.3 Computation of Q.D. for a Continuous Series

Example 6: Calculate quartile deviation and coefficient of quartile deviation from
the following distribution:

Weekly Wages(in ‘000 `) 5-7 8-10 11-13 14-16 17-19

No. of Workers 14 24 38 20 04

Solution:
Table 9.2: Calculation of Quartile deviation and coefficient
of quartile deviation
Weekly Wages No. of Workers (f) Cummulative
(in ‘000 `) Frequency (C.F.)

4.5-7.5 14 14
7.5-10.5 24 38
10.5-13.5 38 76
13.5-16.5 20 96
16.5-19.5 4 100
f = 100

l2 - l1 ( N l
Q1 = l1 + ¨©4 - cf ¸¹
f

N
= 25. Q1 lies in the class of 7.5-10.5
4

l2 − l1 ( N  25 −14
Q1 = l1 +  4 − cf ) = 7.5 +  3 = 8.875 (in `000)
f 24

l2 − l1 ( 3N  75 − 38
Q3 = l1 +  4 − cf ) = 10.5 +  3 = 13.42 (in `000)
f 38

Q3 − Q1 13.42 − 8.875
Q.D. = = = 2.273 (in `000)
2 2
 Measures of Dispersion

Q3 − Q1 13.42 − 8.875
Coeff. of Q.D. = = = 0.21
Q3 + Q1 13.42 + 8.875

Quartile Deviation: An Evaluation

Since Quartile deviation is based only on Q1 and Q3, it means this measure is based
on middle 50% of the data of the series. Thus unlike range, Quartile deviation is
not affected‘ by extreme items as it ignores 25% of the data from the beginning of
the dataset and 25% of the data from the end (data arranged in ascending order).
It can be calculated in case of open-ended distribution. However, it is not based
on all the observations in the data.

INTEXT QUESTIONS 9.3

1. Which of the following is a measure of dispersion?
(a) percentiles
(b) quartiles
(c) inter-quartile range
(d) all of the above are measures of dispersion
2. The inter-quartile range is
(a) the 50th percentile
(b) another name for the standard deviation
(c) the difference between the largest and smallest values
(d) the difference between the third quartile and the first quartile
3. Which of the following limitation of the range is overcome by the inter-
quartile range?
(a) the sum of the range variances is zero
(b) the range is difficult to compute
(c) the range is influenced too much by extreme values
(d) the range is negative
4. Aresearcher has collected the following sample data. The mean of the sample
is 5.
3 5 12 3 2
The inter-quartile range is:
(a) 1 (b) 2 (c) 10 (d) 12
Measures of Dispersion

3. Mean Deviation
Mean deviation (MD) of a series is the arithmetic average of the deviation of
various items from a measure of central tendency (mean, median and mode)

Mean deviation is based on the items of the distribution and is calculated as an

average, on the basis of deviation obtained from either mean, median or mode but
generally from the median.

First we compute deviations of all the items from either mean or median ignoring
plus (+) and (–) signs. They are called absolute values of deviations where the two
parallel bars (ii) indicate that the absolute value is taken. This is also called modulas
value. Then the aggregate of these deviations are divided by the number of
observations this is called mean deviation.

1. Calculation of mean deviation

(i) Arrange the data in ascending order (for calculating of median)

(ii) Calculate median/mean/mode

(iii) Take deviations of items from median/mean ignoring ± signs and denote the
column as |D|

(iv) Calculate the sum of these deviation in case of discrete and continuous series
|D| is multiplied by respective frequency of the item to get f |D|

(v) Divide the total obtained by number of items to get mean deviation

f D
M.D. =
N

(vi) apply the formula to get coefficient of mean deviation

M.D
Coefficient of M.D. = Median / Mean / Mode

Example 6: Calculate mean deviaiton and coefficient of mean deviation from both
mean and median for the following data on the monthly income (in `) of
households

Income (`) 8520 6350 7920 8360 7500

 Measures of Dispersion

Table 9.3: Calculation of mean deviaiton from median

Monthly Deviation from mean Deviations from median
Income (7730) ignoring ± signs (7920) ignoring ± signs
(`) |D| |D|
6350 1380 1570
7500 230 420
7920 190 0
8360 630 440
8520 790 600
X = 38650 |D| = 3220 |D| = 3030

( N +1 
th
and Median = Size of 
2 )
item
X
Mean =
N
( 5 +1 
th
= Size of 
2 )
item
38650
= = 7730
5 = Size of 3rd item
D = ` 7920
M.D. =
N D
Mean Deviation =
3220 N
= = Rs 644
5 3030
= = Rs 606
M.D. 5
Coefficient of M.D. = Mean
M.D.
Coefficient of M.D. =
644 Median
= = 0.083
7730
606
= = 0.076
7920

9.2.3.2 Calculation of mean deviaiton in discrete series

Example 7: Calculate (a) median (b) mean deviation and (c) Coefficient of mean
deviation

Size of item (X) 6 12 18 24 30 36 42

Frequency (f) 4 7 9 18 15 10 5
Measures of Dispersion

Solution.
Table 9.4: Calculation of mean deviaiton from Median
X f cf |D| f|D|
6 4 4 18 72
12 7 11 12 84
18 9 20 6 54
24 18 38 0 0
30 15 53 6 90
36 10 63 12 120
42 5 68 18 90
f = 68 |D| = 72 f|D| = 510

( N +1 
th
Median = Size of  item
2 )

( 68 +1 
th
= Size of  item
2 )

= 34.5th item

f D 510
M.D. = = = 7.5
N 68

M.D. 7.5
Coefficient of M.D. = = = 0.312
Median 24

Calculation of mean deviaiton in continuous series

Example 8: Calculate (i) mean, (ii) mean deviation from mean and (iii) co-efficient
of mean deviation.

Marks 0-10 10-20 20-30 30-40 40-50

No. of students 5 8 15 16 6
Solution: Calculation of mean deviation from mean
 Measures of Dispersion

Table 9.5: Calculation of mean deviaiton from Mean

Marks No of Mid-point m – 25 fd |D| = f|D|
X students f m 10 m – 27

0-10 5 5 –2 –10 22 110

10-20 8 15 –1 –8 12 96
20-30 15 25 0 0 2 30
30-40 16 35 +1 +16 8 128
40-50 6 45 +2 +12 18 108
f = 50 fd = 10 f|D|= 472

fd
X = A+ N c

10
= 25 + 10 = 27 Marks
50

f D 472
M.D. = = = 9.44 Marks
N 50

M.D. = 9.44 = 0.349

Coefficient of M.D. =
Mean 27

Mean deviaiton: An evaluation

Mean deviation ignores the ± signs of the deviation which is mathematically
unsound and illogical. Therefore, this method is non-algebraic. Moreover, it can
not be computed for distribution for open end classes.

INTEXT QUESTIONS 9.4

(i) Calculated mean deviation and co-efficient of mean deviation from median
No of tomatoes per plant 0 1 2 3 4 5 6 7 8 9 10
No of plants 2 5 7 11 18 24 12 8 6 4 3
(ii) Calculate mean deviation from mean
Class: 3-4 4-5 5-6 6-7 7-8 8-9 9-10
Frequency 3 7 22 60 85 32 8
Measures of Dispersion

4. Standard Deviation (S. D.)

Standard deviation is the most important and commonly used measure of
dispersion. It measures the absolute dispersion or variability of a distribution.
Standard deviation is the positive square root of the mean of the squared deviations
of observations from their mean. It is denoted by S.D. or x.

1. Computation of Standard deviation in case of Individual Series

The following four methods are used to calculate the standard deviation:

1. Actual Mean Method

Let X variable takes on N values i.e. X1, X2, ... XN. The standard deviation of these
N observations using actual mean method can be computed as follows:

1. Obtain the arithmetic mean (X) of the given data.

2. Obtain the deviation of each ith observation from X i.e. (Xi – X ). (Note that
(Xi – X ) =0)

3. Square each deviation i.e. (Xi − X)2

4. Obtain the sum in step 3 i.e.  (Xi − X)

N 2
i=1

5. Obtain the square root of the mean of these squared deviations as follows:

 ( X − X)2
Standard deviation (x) =
N
N = Total No. of observation

2. Assumed Mean Method

This method is applied to calculate the standard deviation when the mean of the
data is in fraction. In that case it is difficult and tedious to find the deviation of all
observations from the actual mean by the above method. Thus the deviations (d)
are taken from the Assumed mean (A) and standard deviation is estimated by using
the following formula:
2
d2 (  d 
Standard Deviation (x) = −
N N )
where d = (X – A) i.e. deviation taken from the assumed mean (A)
 Measures of Dispersion

3. Direct Method
The relevance of this method is particularly useful when the items are very small.
To obtain standard deviations, we apply the following formula:

X2
(X)
2
Standard Deviation (x) =
−N

where X = arithemetic mean

(Note: The direct method basically implies taking deviations from zero)

4. Step Deviation Method

In this method we divide the deviations by a common class interval (c) and use the
following formula for computing standard deviation:

d2 (
Standard Deviation (x) = − 2 c
 N d   N )

X−A
where d = ( i.e. deviation taken from the assumed mean and divide by
c )
class interval (c)

Example 9: The wholesale price of a commodity for 6 days in a month of February

2014 is as follows:

Days 1 2 3 4 5 6
Commodity Price(` Per Quintal) 5 15 25 35 45 55

Compute the standard deviation using:

(i) Actual Mean Method

(ii) Assumed Mean Method
(iii) Direct Method and
(iv) Step-Deviation Method
Measures of Dispersion

Solution:
Table 9.6: Calculation of standard deviation

Days Price (X – X) (X – X)2 d= d2 = X2 d = d2

X – 40
(` Per = (X – 30) = (X – 30)2 (X – 40) = (X – 40)2 5
Quintal)

1 5 -25 625 –35 1225 25 –7 49

2 15 -15 225 –25 625 225 –5 25

3 25 -5 25 –15 225 625 –3 9

4 35 5 25 –5 25 1225 –1 1

5 45 15 225 5 25 2025 1 1

6 55 25 625 15 225 3025 3 9

 ( X − X)  ( X − X)2 d d2  X2 d d2

=0 = 1750 = –60 = 2350 = 7150 = –12 = 94

Applying Actual Mean Method

X=
X 180
= = 30 (in Rupees)
N 6

 ( X − X)2 = 1750
Standard deviation (x) = = 17.078 (in rupees)
N 6

Applying Assumed Mean Method:

−60
X =A+ 
d
Here = 40 + = 30
N 6

2
d2 (  d 
Standard Deviation (x) = −
N N )

 d 2 = 2350,  d = −60, N = 6

2350 ( −60 2
 (X ) = − = Rs.17.078
6 6 )
 Measures of Dispersion

Applying Direct Method

X2 −
(X)
2
Standard Deviation (x) =
N

where X = 30,  X 2 = 7150, N = 6

 (X ) = 7150 − (30)2 = Rs.17.078

6

Applying Step Deviation Method

2
d'2 (  d ' 
Standard Deviation (x) = − c
N N )

where c = 5,  d ' 2 = 94, N = 6,  d ' = −12

94 ( −12 2
 (X ) = −  5 = Rs.17.078
6 6 )

Note: The sum of deviations taken from mean is Zero. But the sum of deviations
from a value other than actual mean is not equal to zero

9.2.4.2 Computation of Standard Deviation in case of Continuous Series

In continuous series, the class-interval and frequencies are given. The following
methods are used to compute standard deviation in this case:

1. Actual Mean Method

In this method the following steps are involved:
 Calculate the mean of the distribution.

 Estimate deviations of mid-values from the actual mean i.e. x = m – X .

 Multiply the deviations with their corresponding frequencies to get ‘fx’. [Note
that fx = 0].
 Calculate fx2 values by multiplying ‘fx’ values with ‘x’ values and sum up these
to get  f x 2 .
 Apply the following formula to obtain standard deviation:
Measures of Dispersion

 fx2
Standard deviation (x) =
f

where x = (m − X) i.e. deviation taken from the arithemetic mean (X)

2. Assumed Mean Method

The steps involved in the calculation of standard deviation are as follows:
 Calculate mid-points (i.e. m) of classes.
 Estimate the deviations of mid-points from the assumed mean (A) i.e. d =
m – A.
 Multiply values of‘d’ with corresponding frequencies to get ‘fd’ values (note
that the total of this column is not zero since deviations have been taken from
assumed mean).
 Apply the following formula to calculate standard deviation:

2
 fd2 (  fd 
Standard deviation (x) = −
f f )

where d = (m – A) i.e. deviation taken from the assumed mean (i.e. A)

3. Step Deviation Method

The steps involved in the calculation of standard deviation are as follows:
 Calculate class mid-points (m) and deviations (d) from an arbitrarily chosen
value just like in the assumed mean method. i.e. d= m-A.

m−A
 Divide the deviations by a common factor ‘C’ denoted by d = ( .
c )
 Multiply d values with corresponding f values to obtain fd values.
 Multiply fd values with d values to get fd2 values.
 Obtain fd and fd2 values.
 Apply the following formula.

2
 fd2 (  fd 
Standard Deviation (x) = − c
f f )
 Measures of Dispersion

( m−A
where d =  i.e. deviation taken from the assumed mean and divide by
c )
class interval (c) (or the common factor in case the class intervals are unequal), m
is the mid value of the interval.

Standard Deviation: Interesting Properties

1. The value of Standard Deviation remains same if each observation in a series
is increased or decreased by a constant value i.e. Standard deviation is
independent of change of origin.
2. The value of Standard Deviation changes if each of observation in a series is
multiplied or divided by a constant value i.e. Standard deviation is not
independent of change of scale.
Example 10: A study of 1000 companies gives the following information
Profit (in ` crores) 0-10 10-20 20-30 30-40 40-50 50-60
No. of Companies 10 20 30 50 40 30
Calculate the standard deviation of the profit earned.
(i) Actual Mean Method
(ii) Assumed Mean Method
(iii) Step-Deviation Method

Solution:
Table 9.7: Calculation of standard deviation
Profit No. of m fm d= d = fd fd2 fd fd2
m – 45
(in ` Companies (m – 40)
10
crores)

0-10 10 5 50 –35 –4 –350 12250 –40 160

10-20 20 15 300 –25 –3 –500 12500 –60 180

20-30 30 25 750 –15 –2 –450 6750 –60 120

30-40 50 35 1750 –5 –1 –250 1250 –50 50

40-50 40 45 1800 5 0 200 1000 0 0

50-60 30 55 1650 15 1 450 6750 30 30

6300 d = –60 d = –9 –900 40500 –180 540

Measures of Dispersion

Applying Actual Mean Method

 fx2
Standard deviation (x) =
f

 fx2 = 36000,  f = 180

36000
(X ) = = 14.142(in rupees crores)
180

Applying Assumed Mean Method:

2
 fd2 (  fd 
Standard deviation (x) = −
f f )

 fd2 = 40500,  f d = −900,  f = 180, d = X − 40

40500 − ( −900 2
(X ) =  = 14.142(in rupees crores)
180 180 )

Applying Step Deviation Method:

2
 fd'2 (  fd '
Standard deviation (x) = − c
 f  f )

m − 45
d' = ,  fd'2 = 540,  fd ' = −180,  f = 180, c = 10
10

540 − ( −180 2
(X ) =  10 = 14.142(in rupees crores)
180 180 )

Example 11: The following table shows the daily wages of a random sample of
construction workers. Calculate its mean deviation and standard deviation.
 Measures of Dispersion

Daily Wages (`) Number of Workers

200 - 399 5
400 - 599 15
600 - 799 25
800 - 999 30
1000 - 1199 18
1200 - 1399 7
Total 100

Solution:
Table 9.8: Computation of mean deviation
Daily Wages Number of Class Mark fm fi |m –X|
(`) X Workers (f) (m) = fi |m – 823.5|
200 – 399 5 299.5 1497.50 2,620
400 – 599 15 499.5 7492.50 4,860
600 – 799 25 699.5 17487.50 3,100
800 – 999 30 899.5 26985.00 2,280
1000 – 1199 18 1,099.5 19791.00 4,968
1200 – 1399 7 1,299.5 9096.50 3,332
Total 100 82350.00 21,160
fi m − X 21,160
Mean deviation = = = 211.60 (`)
fi 100
Table 9.9: Computation of Standard deviation
Daily Wages Number of Class Mark fi (m – X)2
(`) Workers (M.V.)

200 – 399 5 299.5 1, 372,880

400 – 599 15 499.5 1,574,640
600 – 799 25 699.5 384,400
800 – 999 30 899.5 173,280
1000 – 1199 18 1,099.5 1,371,168
1200 – 1399 7 1,299.5 1,586,032
Total 100 6,462,400

6462400
Standard deviation = = 254.21 (Rupees)
100
Measures of Dispersion

INTEXT QUESTION 9.5

1. Sona, Karina, Omar, Mustafa and Amie obtained marks of 6, 7, 3, 7, 2 on a

standardized test respectively. Find the standard deviation of their scores.

9.2.4.3 Comparison of the variation of two series using standard deviation

The values of the standard deviations cannot be used as the basis of the comparison
mainly because units of measurements of the two distributions may be different.
The correct measure that should be used for comparison purposes is the Coefficient
of Variation (C.V.) given by Karl Pearson:

X
C.V. = 100
X

X = S.D. of var iable X, X = mean of var iable X

Example 12: The following table shows the summary statistics for the daily wages
of two types of workers.

Worker’s Type Daily Wages

Mean Standard deviation

I ` 100 ` 20
II ` 150 ` 24

Compare these two daily wages distributions.

Solution:
Table 9.10: Calculation of coefficient of variations
In comparison Distribution Reason

Average magnitude II > I XII = 150  X I = 100

20
Variation I > II CV = 100 = 20%  CV
I II
100
24
= 100 = 16%
150
 Measures of Dispersion

INTEXT QUESTIONS 9.6

1. The hourly wages of a sample of 130 system analysts are given below:
mean = 60 range = 20
mode = 73 variance = 324
median = 74
The coefficient of variation equals
(a) 0.30% (b). 30% (c) 5.4% (d) 54%
2. The measure of dispersion that is influenced most by extreme values is:
(a) the standard deviation
(b) the range
(c) the inter-quartile range
3. The descriptive measure of dispersion that is based on the concept of a
deviation about the mean is
(a) the range (b) the inter-quartile range
(c) the absolute value of the range (d) the standard deviation
4. The numerical value of the standard deviation can never be
(a) zero (b) negative (c) one
5. Aresearcher has collected the following sample data. The mean of the sample
is 5.
3 5 12 3 2
The Standard deviation is
a. 8.944 b. 4.062 c. 13.2 d. 16.5
Answer: b
And
The Coefficient of Variation is
a. 72.66% b. 81.24% c. 264% d. 330%
Answer: b

9.2.5 Lorenz Curve

Lorenz curve is the graphical method of studying dispersion. Lorenz curve is the
cumulative frequency curve showing the distribution of a variable such as
population against any independent variable such as income or area settled. If the
Measures of Dispersion

distribution of the dependent variable is equal, the plot will show as a straight, 45°
line. Unequal distributions will yield a curve. The gap between this curve and the
45° line is the inequality gap. The farther the curve from this 45o line, the
greater is the variability present in the distribution. Lorenz curve is used to see
the degree of concentration of income or health. For example, it may show
top 25% of population accounts for 70% of income or bottom 25% of population
has only 5% of income (see figure 9.1).

Last 25%
has 70% of
Cumulative % of Income

national
income

Third 25%
has 15% of
national
income
Second 25%
has 10% of
national
income
First 25%
has 5% of
national
income

Cumulative % of population

Fig. 9.1

Steps involved in drawing Lorenz Curve:

The drawing of Lorenz curve requires following steps:
1. Find cumulative totals of variables. In case of continuous variable calculate the
cumulative totals of mid-points.
2. Estimate cumulative frequencies.
3. Express the cumulative mid-points and frequencies into percentages by taking
each of the sum total as 100.
4. Take the cumulative percentages of the variable on Y axis and cumulative
percentages of frequencies on X-axis. Each axis will have values from ‘0’ to
‘100’.
5. Draw a line joining Co-ordinate (0, 0) with (100,100). This is called the line
of equal distribution.
6. Plot the cumulative percentages of the variable with corresponding cumulative
percentages of frequency. Join these points to get the Lorenz Curve.
 Measures of Dispersion

WHAT YOU HAVE LEARNT

 The important measures of dispersion are:
i) Range
ii) Quartile deviation or Semi-Inter quartile range.
iii) Standard deviation
iv) Lorenz Curve
 Range (R) is the difference between the largest (L) and the smallest value (S)
in a distribution i.e. Range (R) = L – S
 The Coefficient of Range is the relative measure of the range and is given by:
L −S
L +S
Q3 − Q1
 Quartile Deviation (Q.D) is given by Q.D. =
2

Q3 − Q1
 The Coefficient of quartile deviation is given by Coeffof Q.D. =
Q3 + Q1

 Standard deviation is the most important and commonly used measure of

dispersion It is denoted by S.D. or x.
 Standard deviation in case of Individual Series is given by four methods:

(i) Actual Mean Method

 ( X − X)2
Standard deviation (x) =
N

N = Total No. of observation

(ii) Assumed Mean Method

2
d2 (  d 
Standard deviation (x) = −
N N )

where d = (X – A) i.e. deviation taken from the assumed mean (A)

Measures of Dispersion

(iii) Direct Method

X2 −
(X)
2
Standard deviation (x) =
N

where X = arithemetic mean

(iv) Step Deviation Method

2
 d '2 (  d ' 
Standard deviation (x) = − c
N N )

(X−A
where d = i.e. deviation taken from the assumed mean and
 c )
divide by class interval (c)
 Standard Deviation in case of Continuous Series is given by

(i) Actual Mean Method

 fx2
Standard deviation (x) =
f

where x = (m − X) i.e. deviation taken from the arithemetic mean (X)

(ii) Assumed Mean Method

2
 fd2 (  fd 
Standard deviation (x) = −
f f )

where d = (m – A) i.e. deviation taken from the assumed mean (i.e. A)

(iii) Step Deviation Method

2
 fd2 (  fd 
Standard deviation (x) = − c
f f )

( m−A
where d =  i.e. deviation taken from the assumed mean and
c )
divide by class interval (c) (or the common factor in case the class
intervals are unequal, m is the mid value of the interval.
 Measures of Dispersion

 Standard deviation is independent of change of origin but not independent of

change of scale.
 The coefficient of variation (C.V.) is the relative measure of dispersion which
is used for the comparison of variability of two or more distributions. It is
given by:

X
C.V. = 100
X
X = S.D. of var iable X, X = mean of var iable X
 Lorenz curve is the graphical method of estimating dispersion.

TERMINAL EXERCISES
Range
1. The following are the prices of shares of A B Co. Ltd. from Monday to
Saturday:
Days Price (in `) Days Price (in `)
Monday 200 Thursday 160
Tuesday 210 Friday 220
Wednesday 208 Saturday 250

Calculate range and its coefficient

2. Find the range of given data
108, 107, 105, 106, 107, 104, 103, 101, 104
3. Find the value of range of frequency distribution
Age in years: 14 15 16 17 18 19 20
No. of students : 1 2 2 2 6 4 0
4. Calculate the range for the distribution given below
Height in cms 150 151 152 154 159 160 165 166
No. of Boys 2 2 9 15 18 10 4 1
5. Find the range of the following data
Profit (in ‘000 `): 0-10 10-20 20-30 30-40 40-50
No. of firms 0 6 0 7 15
Measures of Dispersion

6. Find the range of the following distribution

Class Interval 10-20 20-30 30-40 40-50 50-60
Frequency 8 10 15 18 19

Quartile Deviation
7. Calculate the QD for a group of data,
241, 521, 421, 250, 300, 365, 840, 958
8. From the following figures find the quartile deviation and its coefficient:
Height (cms.): 150 151 152 153 154 155 156 157 158
No. of Students: 15 20 32 35 33 22 20 12 10
9. Using quartile deviation, state which of the two variables –Aand B is more variable:
A B
Mid-Point Frequency Mid-Point Frequency
15 15 100 340
20 33 150 492
25 56 200 890
30 103 250 1420
35 40 300 620
40 32 350 360
45 10 400 187
450 140
10. Find the quartile deviation from the following table:
Size: 4-8 8-12 12-16 16-20 20-24 24-28 28-32 32-36 36-40
Frequency: 6 10 18 30 15 12 10 6 2
11. Calculate the coefficient of quartile deviation from the following data:
Class Interval Frequency
10–15 4
15–20 12
20–25 16
25–30 22
30–40 10
40–50 8
50–60 6
60–70 4
8
 Measures of Dispersion

Standard deviation

12. Determine the standard deviation of the following student test results
percentages.

92% 66% 99% 75% 69% 51% 89% 75% 54% 45% 69%

13. Calculate the coefficient of variation for the following data set.

The price (in `), of a stock over five trading days was 52, 58, 55, 57, 59.

14. The frequency table of the monthly salaries of 20 people is shown below.

Salary (in `) Frequency

3500 5

4000 8

4200 5

4300 2

(a) Calculate the mean of the salaries of the 20 people.

(b) Calculate the standard deviation of the salaries of the 20 people.

15. The following table shows the grouped data, in classes, for the heights of 50
people.

Height (in cm) – classes Frequency

120  130 2

130  140 5

140  150 25

150  160 10

160  170 8

a) Calculate the mean of the salaries of the 50 people.

b) Calculate the standard deviation of the salaries of the 50 people.

Measures of Dispersion

16. The following is the frequency distribution for the speeds of a sample of
automobiles traveling on an interstate highway.
Speed Miles per Hour Frequency
50 – 54 2
55 – 59 4
60 – 64 5
65 – 69 10
70 – 74 9
75 – 79 5
35

Calculate the mean, and the standard deviation of speed.

17. In 2012, the average age of workers in a company was 22 with a standard
deviation of 3.96. In 2013, the average age was 24 with a standard deviation
of 4.08. In which year do the ages show a more dispersed distribution? Show
your complete work and support your answer.
Therefore the year 2012 shows a more dispersed distribution.
18. The following is a frequency distribution for the ages of a sample of employees
at a local company.

Age (in years) Frequency

30 – 39 2
40 – 49 3
50 – 59 7
60 – 69 5
70 – 79 1

(a) Determine the average age for the sample.

(b) Compute the standard deviation.
(c) Compute the coefficient of variation.
19. The population change between 1990 and 2000 for several small cities are
shown below.
 Measures of Dispersion

City Population Change

(number of residents)

A 3083
B 1466
C -461
D 1113
E -11
F 395
G 3290
H 437

For the above sample, determine the following measures.

(a) The mean
(c) The standard deviation
(d) The median

Answers:
ANSWERS TO INTEXT QUESTIONS

9.1
1. (c)
2. (c)

9.2
1. 140

9.3
1. (c)
2. (d)
3. (c)
4. (b)
Measures of Dispersion

9.4
1. Median = 5, M.D. = 1.68
2. M.D. 0.915, Coefficient of M.D. = 0.336

9.5
1. 2.1 marks

9.6
1. (b)
2. (b)
3 (d)
4. (b)
5. Standard deviation (b) C.V