CHAPTER 6

DISPERSION

ONE SHOT REVISION

DISPERSION
The degree to which numerical data tend to spread about an
average value is called the dispersion (or variation of the data). ie,
Dispersion is the degree of scatter or variation of the variable
about a central value.

① 2 3 => mean =
3 =R
=
4 -2
=
2
, , 4
1= 4
② 3 =R 5
=

I,
-

3 => mean
=

5
,

③ 0. 3, 6 = mean =
3 =R
= 6 - 0 =

z
6
Several measures of variability or dispersion are,

• Range

• Quartile Deviation

• Mean Deviation

• Standard Deviation
 Range
It is the difference between the highest value (H) and the lowest
value (L) in a set of data. In symbols,

R=H-L

• Higher value of range implies higher dispersion and vice versa.

Find the range of the data.
&
25, 30, 45, 100, 120,&
20

ans H = 120

L = 20

Range H-L
=

120 - 20
=

Range = 100

=
Quartile Deviation
Quartile deviation is also called semi-inter quartile range.

Interquartile Range = Q31

-Q

Quartile Deviation = 93 -
QI
2

Q = First quartile
I

Q = Third Quartile
3
 Discrete Continuous
Raw Data Frequency Frequency
Distribution Distribution

Arrange the data in ascending order of magnitude

Find QI and Q3

QD Q3 Q ,
-

2
? Find the interquartile range and quartile deviation.
2,5,8,1,6,9,7 n= 7

ans Order => 1 , 2 5 6 7 8, 9

Ascending , , , ,

th
Q =

(th value
=
value
nd value =
2
=
= 2

93 =

3 (th value =
3 (thvalueIthvalue =

=
3X2

Q3 S -
=
 Q 2

Interquartile Range Q3-Q ,

=

S
Q3
=

=
8 2 -

= 6

Quartile Deviation =
Q3-Q,
2

=
82
2

-6
2

Q . D =

=
3
Mean Deviation (MD)
MD is a measure of dispersion which depends on all values in
a series.
Mean deviation is the arithmetic mean of absolute deviation
of observations from an assumed average.

M D x A
A is
any average
-
.
=

A = mean

A-median
A = mode
Calculate the mean deviation about mean for the given data.
1 2 3 4, 5 n 5 mean-i
A
=
, , , =

and m D.
about
mean=E
2x +5
m =
-
= 1+ 2+ 3 + 4
-
U
S

=
n = 3
z
 (2) =
2

x (x) T =
3
(2) = 2

I 11-31 = 2

2 (2-3) = 1
E(x-5) =
6
=
13-3) 0

i
=

about
1
14 3
=
-

m D
.

15-31 = 2
mean
15 6
=

5
M D
.
=
1 .
2

=
Mean Deviation for a Discrete Frequency and
Continuous Frequency Distribution

Mean Deviation f(x-Al A is any

=

,
N
average
Find Mean Deviation about median.

x 12 3 4

f4786

ans
median()th value =
2
=
N = Ef = 25
-ithvalue
-
x f C .
f

i
I 4

2 T
Median = 3

8 19
3

4 6 25

m D =

E-A
.
 median = 3

R ↓ (x-median) f(x-median Ef(x-median)

11-31 2 S
T
I =

= 21
2 12-31 1 =
T
13-31 0 N = 25
8
=

O
3

4 6 (4=
3) = 1 6

Total 25 21

median-median
m .

D about -

25

-
Standard Deviation
Standard deviation (SD) is defined as the positive square root
of the mean of squares of deviations from the arithmetic mean.
• Denoted by 5
• It cannot be negative.
• It is the best measure of dispersion.
A relatively small standard deviation
indicates high degree of uniformity
in the data with not much variation
of individual values from their mean.

If all the observations are equal, the SD

is equal to zero.
Properties
1. The minimum value of SD is zero.
2. The sum of squared deviation is minimum when
taken about mean.
Standard Deviation for Raw Data
SD , v =
(x j)2
-

r =
x -
U
 Find standard deviation for the given data.
2 3 5 7 10
, , , ,

and r
Ex (
= n =
5
-

Ex =
187
<
=
17-15 4) .
i =
5
.

4
5

j= 2 .
87
=
37 14 . -
29 16.

87
22
.
= =
8 24 .
 2 se2
n =
x n =
5

a
2
H
3
5 25 - 27
-

T 49 5

10 100 n =
5 .

4
Ex = 21 En 187 -
Standard Deviation for Discrete Frequency Distribution
and Continuous Frequency Distribution

S D
.

,
v
= -)

N = Ef

=
Variance

Variance = -2

SD =
variance = r
Absolute measures of dispersion Relative measures of dispersion
• Range • Coefficient of Variation
• Quartile deviation • Coefficient of quartile
• Mean deviation deviation
• Standard deviation
Measure of dispersion which is
Measures the variability or independent of unit of
deviation of values in a data and measurement.
expressed in the original unit of • It is the ratio of absolute
that data. measures of dispersion to
an average with which the
measure of dispersion is
computed.
Coefficient Of Variation (CV)
This measure is used to compare the consistency or stability
between two or more sets of data.

Coefficient of variation, CV = - X 100

I

• This measure is free from the unit of original measurements.

• Less CV, more consistency.

• Greater the CV, less consistency.
 A B
- -

SD .
= += 5 S D .
=
+ =
4
20 Mean 25
mean =
= = x =

C Ve x100
C V= =
.

x100

= =
CVI =

=
25 CV2
 A = C V
-
= 25

B = ( V . = 16-

B is more consistent
 Coefficient of Quartile Deviation
It can be used to compare the degree of variation in
different situations.

Deniation
Coefficient of Quartile
=
 Find the coefficient of quartile deviation.
n 7
1 3
=

4, 7 10 12 8
, , , , ,

and , 411 00
Ascending Order = 1
,
8, ,
12

th nd value
q,
= value
= Q, =
3

03 = 3
(th value =

3(7) =
3x2 =

Q3
6th
=
value
10
Co-efficient of Quartile Deviation
= Q Q
,
=
3

Q3 + Q 1 Q3 = 10

=
=
= 0 .
538
=
Covariance
It is a measure of strength of linear relationship between
two variables. It indicates whether the variables are
positively or negatively related in a bivariate distribution.

Cov(x , y) =
z(x 5)(y
-
-

j)
R

=-x

number observations
n =
of
 Find covariance.

x 436
M
2547
y

and
Cov(y) =

y-z Exy
=
76

n =
↑
=
16 -

(3 .

5x 4 .

5) I = 3 5 .

4 4 5
j
.
=

=
19-15 .
75

Corby) =
3 25
.

-
 n 4
=

·
= Ex
T

y=
Exy 76
=
=