Measures of

BBA 104
dispersion
1
 Outline
• Importance of the concept of variability
• Measure the spread or dispersion,.
• Know the requisities of an ideal measure of dispersion
• Identify and calculating various methods of dispersion.

BBA 104
2
Dispersion
• Dispersion is the measure of the variation of the items. It is
the degree of the scatter or variation of the variables about a
central value. It give an average of the differences of various
items from an average.

BBA 104
3
Significance
• To determine the reliability of an average.
• To serve as a basis for the control of the variability.
• To compare two or more series with regard to their variability.
• To facilitate the use of other statistical techniques.

BBA 104
4
Classification of measures of
Dispersion

BBA 104
Absolute Relative
Measure Measure

5
Absolute Measure
• These are described by a number or value to represent the
amount of variation or differences among values in a data set.
• Such a number is expressed in the same unit of measurement
as the set of values in the data such as rupees, grams,etc.
• Are used to compare two or more sets of data provided the

BBA 104
variable values are expressed in same unit of measurement.

6
Relative Measure
• These are described as the ratio of a measure of absolute
variation to an average and is termed as coefficient of
dispersion where coefficient means a pure number that is
independent of the unit of measurement.
• It is important to note that while calculating relative

BBA 104
dispersion the average used as base should be the same one
from which the absolute deviations were meaured.

7
Measures of Dispersion

Range Interquartile
Deviation

BBA 104
Mean Standard
Deviation Deviation

Lorenz
Curve 8
Range
The range is the most simple measure of dispersion and is based
on the location of largest and the smallest values in the data.
It is defined as the difference between the largest and lowest
observed values in a data set.

BBA 104
9
 BBA 104
10
Properties:
1.Greater the range, larger is the variability among the data set.
2.If a constant is added (or substracted) to all the values, range is not
affected ,hence, it is independent of origin.
3.If all the values are multiplied (or divided) by a constant range is

BBA 104
multiplied (or divided)by the same constant. Hence, if scale changes,
range also changes.
4.It is used in quality control and in construction of control
charts.example: measurement of risk in investment via stock price
quotations,etc.

11
Questions
Calculate range and its coefficient.

Serial 1 2 3 4 5 6 7 8 9
No,

NPV 11 9 22 8 24 17 10 12 13

BBA 104
Earnings 5-10 10-15 15-20 20-25 25-30 30-35 35-40 40-45

No. of 25 28 32 21 17 14 8 6
companies 12
For ungrouped data:
Range = H – L
Where , H refers to highest value of the series
L refers to lowest value of the series.
For grouped data :
Range = difference between upper class limit of the last class and the
lower class limit of the first class.
Coefficent of Range = H – L

BBA 104
H+L

13
Merits & Demerits
Merits:
1.Easy to calculate
2. Not affected by frequencies
Demerits :

BBA 104
1.Based on highest and lowest values ,thus ignores other values
of the series.
2. It is influenced by extreme values.
3. It is not suitable for open –ended distribution.
4.It cannot be used for further algebraic manipulations.

14
Inter –Quartile Range
It measures the range of the middle fifty percent values of the
data set so as to minimise the influence of outliers(extreme
values)in the calculation of range.
It overcomes the limitations of another measure of variation i.e.
range.

BBA 104
It is an absolute measure of variation.
It is a measure of values in the data set between third quartile &
first quartile.

15
 BBA 104
16
 BBA 104
17
 • Inter- Quartile Range = Q₃ - Q₁
Question : Calculate IQR

BBA 104
Marks 10 20 30 40 50 60 70 80 90

No of 90 50 90 60 54 40 70 15 30
stude
nts

18
Quartile Deviation
It is the average of the inter quartile range. It is an absolute
measure of dispersion.
Quartile Deviation = Q₃ - Q₁
2

BBA 104
Coefficient of Quartile Deviation = Q₃ - Q₁
Q₃ + Q₁

19
Questions
Amount 25 50 75 100 200 300
of tax
No. of 135 120 90 60 55 40
traders

BBA 104
Daily 4-9 10-14 15-19 20-24 25-29 30-34
collection
(in
thousands
)
No. of 66 72 86 53 49 30
days
20
Mean Deviation
Average of sum of absolute deviations from Median (or Mean) is
called Mean Deviation.
It is also known as Mean Absolute Deviation.
In general, it is better to take deviations from Median as it is the

BBA 104
least in this case.

21
Mean Deviation for Individual series:
M.d. = ∑|D|
n
Where : |D| refers to |x- Mean|
or |x- Median|

Mean Deviation for Discrete or Continous Series

M.d. = ∑f|D|
∑f

BBA 104
Where : |D| refers to |x- Mean|
or |x- Median|

Coefficient of Mean Deviation :

CMD = M.D.
M or 𝑋( from where deviations are taken)
:
22
Runs 6 10 12 20 24 36 40 48
Score
d
No. of 7 11 14 36 16 9 5 1
Player

BBA 104
s

Class 0-10 10-20 20-30 30-40 40-50 50-60

Frequency 6 7 15 16 4 2

23
Merits & Demerits
Merits:
1. It is based on all observations.
2. It is easy to compute as negative figures are also considered
as positive.

BBA 104
3. It is not affected by extremely large values.
Demerits:
1. The algebraic signs are ignored.
2. It cannot be used for further algebraic treatments.

24
Standard Deviation
It is the positive square root of the averages of the squares of the
sums of deviations, deviations taken from mean. It represents the
variability of data from the average. Standard deviation is a
mathematical tool to help us assess how far the values are
spread above and below the mean.

BBA 104
A low standard deviation indicates that the data points tend to
be close to themean (also called the expected value) of the set,
while a high standard deviationindicates that the data points are
spread out over a wider range of values.

25
Features of Standard
Deviation
• If a constant is added to (or substracted)to all variables, S.D. is
not affected. Hence, it is independent of origin.
• If all the variables are multiplied (or divided) by a constant ,
S.D. is also multiplied (or divided) by the constant. Hence, it is
not independent of scale.

BBA 104
• Incorrect S.D. can be corrected.
• Combined S.D. can be calculated.
• Value of S.D. is always greater than or equal to zero

26
 • In case of normal distribution :
68.27% of the items will fall within 𝑥 ± σ
95.45% of the items will fall within 𝑥 ± 2σ
99.73% of the items will fall within 𝑥 ± 3σ

BBA 104
27
Days 1 2 3 4 5 6 7

Commo 240 260 270 245 255 286 264

dity
Price

BBA 104
Profit 0-10 10-20 20-30 30-40 40-50 50-60
No. of 8 12 20 30 20 10 28
compani
es
Combined standard
Deviation

BBA 104
29
 • Where :
d₁ = 𝒙₁₂ - 𝒙₁
d₂ = 𝒙₁₂ - 𝒙₂

BBA 104
𝒙₁₂ = n₁x₁+n₂x₂
n₁+n₂

30
Coefficient of variation
The relative measure of dispersion is Coefficient of Variation. It is
very useful measure for :
1. Comparing two or more data sets expressed in different
units of measurement.
2. Comparing data sets that are in same unit of measurement

BBA 104
but the mean values of data sets in a comparable field are
widely dissimiliar( mean wages received per month by top
management personnel and labour class personnel of the
same organisation.)
It measures the standard deviation relative to the mean in
percentages. It indicates how large the standard deviation is in
relation to the mean .
CV = Standard Deviation
31
Mean *100
• Lower the CV , more consistent or homogeneous.
• Higher the cv , less consistent or less homogeneous.

BBA 104
32
 He
This method of dispersion was developed Dr. Max o. Lorenz .It is
• Stepsmethod
graphic for Preparing
and isLorenz
knownCurve
as Lorenz Curve .It is a cumulative
frequency curveof.The
1. The values procedure
the variable are of preparing
added this curve
and expressed is as
in % under.
form.
2. The frequencies of the series are added and expressed in %
form.
3. Both percentage of variables and frequencies are converted
into cumulative form.
4. The cumulative % of variables are represented on x-axis and
• cumulative % of frequencies are represented on y-axis .
5. The diagonal line (O) is joined with 100. This line is called line
of distribution.
6. The cumulative % of variables and cumulative % frequencies
are

p lo ttedon graph.The curve soBBfAo10r6medby joining these plotted

12 /23/ 20 1 3 17
1.It is a simple measure of dispersion.
2.The quantity of dispersion can be known merely
by looking at the Lorenz Curve.
3.The variability between two series can also be
compared by this method .
4.Lorenz Curve can be successfully drawn in
distributions pertaining to profit , wages,
distribution of income and wealth , etc.
12/23/2013 BBA 106 3
4
I. We can know the quantum of dispersion in
the series by a mere look on the graph but it
cannot be measured numerically.
II. The process of preparing Lorenz Curve is
complex .
III. It cannot be prepared for individual series.

12/23/2013 BBA 106 3

5
Summary
After studying the student is able to answer:
• What is the concept of variability
• How to measure the spread or dispersion,.
• What are the requisities of an ideal measure of dispersion

BBA 104
• Calculating dispersion using various methods such as range,
mean deviation ,standard deviation,etc

36