Lesson# 4

measure of Dispersion

Department of Statistics
FC College University, Lahore

1
  You have learnt various measures of central tendency.
 Measures of central tendency help us to represent the entire mass
Preview of the data by a single value.
 Can a central tendency describe the data fully and adequately?
 In order to understand it, let us consider an example ,

2
  Here we observe that in both groups the mean of the data arethe
Example same namely, 65
I. In group A, the observations are much more scattered from the
mean
II. In group B, almost all the observations are concentrated around
The daily income the mean

of the workers in
two factories are : Thus, there arises a need to differentiate between the groups. We need
some other measures which conern with the measure of scatteredness
Factory A: 35 45 50 65 70 90 100 (or spread).
Fcatory B: 60. 65 65 65 65 65 70
To do this, we study what is known as measures of dispersion

3
  The term dispersion is used commonly to mean scatter, deviation,
fluctuation, spread or variability of data.

Meaning of  The degree to which the individual values of the variate scatter
away from the average or central value, is called dispersion.
Dispersion  Dispersion is used to denote a lack of uniformity in the item
values of a given variable.
Dispersion/
Variation
Everywhere
 Measures of Dispersion

A quantity that measures

the variability among the
data, or how the data set
dispersed about the
average, known as
Measures of dispersion,
scatter, or variations
 Dispersion can be Large or Small

 The more similar the

scores are to each other,
the lower the measure of
dispersion will be.
 The less similar the
scores are to each other,
the higher the measure
of dispersion will be.
  An average, such as mean or the median only locates the center of
the data.
 An average does not tell us anything about the spread of the data.
Why Study  Determine the reliability of an average.
Dispersion?  To compare the variability of two or more series.
 To facilitate the use of other statistical measures.
 Types of dispersion

Absolute
Relative dispersion
dispersion

9
 ABSOLUTE DISPERSION RELATIVE DISPERSION
Types of  These measures gives us an idea  These measures are calculated for
about the amount of dispersion in the comparison of dispersion in
Dispersion a set of observations. two or more than two sets of
observations.
 These quantities measures the
dispersion in the same units as the  These measures are free of the
units of original data. units in which the original data is
measured.
 Absolute measures cannot be used
to compare the variation of two or  These measures are a sort of ratio
more series/ data sets. and are called coefficients

10
 Range Coefficient of Range

Absolute measures Relative measures of

of dispersion dispersion
Interquartile range/ Coefficeint of Quartile
Quartile deviation deviation

Measures of
dispersion

Mean deviation Coefficient of Mean

deviation

Standard deviation Coefficient of Variation

 Difference between Absolute and Relative Measures of Dispersion

ABSOLUTE MEASURES RELATIVE MEASURES

Has the variable unit.

Not expressed in terms of percentage. Has no unit.
Calculated from original data. Expressed in terms of percentage.
No use of ratio Calculated from absolute measures
Cannot be used to compare the variation of Use of ratio.
two or more series. Can be used to compare the variation of two or
No change in variables and with the absolute more series.
measures. Changes in variables with relative measures.
1. Range

13
 Range is defined as the difference between the maximum and the
minimum observation in the data set.
The Range If denotes the minimum observation, then range is defined as:
(Absolute Range= -
Measure) Where,
= maximum observation
= minimum observation
  Example 1: Cheryl took 7 math tests in one marking period. What
is the range of her test scores?
89,  73,  84,  91,  87,  77,  94
Example-1  Range = 94-73=21
  The range of these test scores is 21 points

15
 It is a relative measure of dispersion and is based on the
Co-efficient of value of range. It is also called range coefficient of
dispersion. It is defined as
Range
Co-efficient of Range =
(Relative
The range is standardized by the total .
Measure)
 Example -2
A testing lab wishes to test two experimental brands of outdoor paint to see how long each will last before fading.
The testing lab makes 6 gallons of each paint to test. Since different chemical agents are added to each group and only six cans
are involved, these two groups constitute two small populations. The results (in months) are shown
Find the range for the paints.

For brand A, the range is For brand B, the range is

Range=60 -10 =50 months R =45 -25 =20 months
Co-efficient of Range = Co-efficient of Range =

We see that the coefficient of range for Brand A is greater than that of brand B.
Thus, there is greater dispersion or variation in Brand A. The time duration for paint fading in Brand B outdoor paints
are more stable than Brand A. 
Example 3: Following are the
wages of 8 workers of a
factory. Find the range and
the co-efficient of range.
Solution:
Wages in $.
1400,1450,1520,1380,1485,149 Largest value Xm= 1575 and
5,1575,1440
Smallest value = X0= 1380
Range = Xm – X0 = 1575-1380
= $195
Co-efficient of Range =
=
Example 4: Set A contains the
marks of 5 students in Set B: English
Set A: Mathematics
Mathematics out of 25 marks 30,35,40,45,50
and group B contains marks of 10,15,18,20,20
the same students in English Range = Xm – X0
out of 100 marks. Compare the Range = Xm – X0
= 50-30 = 20
data with Coefficient of Range. = 20-10 = 10
Co-efficient of Range
Co-efficient of Range = = =
= ==
= 0.33
= 0.25

We see that coefficient of range for set A is greater than that of set B. Thus there is greater dispersion in
set A.
Merits and Demerits of
Range
MERITS DEMERITS

 Value depends entirely upon the

extreme values.
 Easy to compute.
 Influenced by sample size.
 Simple and easy to understand.
 Very sensitive to outliers.
 Left itself to algebraic treatment.
 Not based on all the observations.
 Insensitive to the distribution of
scores within the two extremes.
 PRACTCIE QUESTIONS

(a)1. Attacks by stinging insects, such as bees

or wasps, may become medical emergencies 2. The outdoor temperature, in degrees
if either the victim is allergic to venom or Fahrenheit, is recorded at noon for ten
multiple stings are involved. The following consecutive school days. The measurements
data give the number of patients treated were as follows:
each week for such stings in a large regional 370, 420, 380, 260, 350, 250, 380, 340, 280, 230
hospital during 13 weeks last summer. What is the Range of temperature?
Also find Coefficient of Range.
1 5 2 3 0 4 1 7 0 1 2 0 1
Compute the range
2. Quartile deviation

22
Quartile It is based on the lower quartile Q1 and the upper quartile Q3. The
Deviation difference Q3- Q1 is called the interquartile range. The
difference Q3- Q1 divided by 2 is called semi-interquartile range or
(Absolute the quartile deviation.
Dispersion) Quartile Deviation = =
The quartile deviation is a slightly better measure of absolute
dispersion than the range. But it ignores the observations on the
tails.
Coefficient of
Quartile A relative measure of dispersion based on the quartile deviation is called the
Deviation coefficient of quartile deviation. It is defined as

(Relative Coefficient of Quartile Deviation =

=
Measure) It is a pure number free of any units of measurement. It is used for comparing
the dispersion in two or more sets of data.
Example 5: 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,
1720,1600.1470,1750,1885. 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
= Value of
= Value of

=5thObservation+0.25 (6thObservation- 5thObservation)

= 1240 +0.25 (1320-1240) = 1240+20=1260kg
Example 5: Cont....

= Value of 3
= Value of 3
= Value of
=15thObservation+0.75(16thObservation-15thObservation)
= 1750+ 0.75(1755-1750) = 1750+3.75 = 1753.75

Quartile Deviation= =246.875 kg

Coefficient of Quartile Deviation==
= = 0.1638
Example 6:

The Thomas Supply Company Inc. is a distributor of gas-powered

generators. As with any business, the length of time customers take to
pay their invoices is important. Listed below, is the time, in days, for a
sample of The Thomas Supply Company Inc. invoices.
41 13 20 26 27 31 34 34 34 54 35 36 37 82 41 41 38
45 47 47 47 50 51 53 13 56 62 67 34 23
Find Quartile deviation and Coefficient of Quartile Deviation
Example 6: Cont.….

Solution:
First we have to arrange our data set in ascending or descending order
13 13 20 23 26 27 31 34 34 34 34 35 36 37 38
41 41 41 45 47 47 47 50 51 53 54 56 62 67 82

First Quartile == Value of

= Value of

= 7th Observation + 0.75 (8th Observation – 7th Observation)

= 31 +0.75 (34-31) = 33.25
Third Quartile == Value of 3
= Value of 3
 = Value of
=23rd Observation+0.25(24th Observation-23rd Observation)
= 50 +0.25 (51-50)
= 50+0.25
=50.25
Quartile Deviation =
Quartile Deviation = = 8.5 days
Coefficient of Quartile Deviation== =
= 0.20359
Video Aid:
https://youtu.be/-JxAeFwgQh4
Example 7: The following data refers to the dividend (%) paid by two companies A and B over the last 5 years.
Compare the two series by coefficient of Quartile Deviation.

Set B:
Set A:
nB=5
nA=5
5,8,10,12,15
4, 8,10,11 15
= Value of
= Value of
= Value of
= Value of

=1 obs+0.5(2 obs-1 obs) = 4+0.5(8-4) = 6

st nd st =1st obs+0.5(2nd obs-1st obs)= 5+0.5(8-5) = 6.5
= Value of3 = Value of3
= Value of = Value of
= 4th obs+0.5(5th obs-4th obs)=11+0.5(15-11)=13 = 4th obs+0.5(5th obs-4th obs) =12+0.5(15-12)=13.5
Quartile Deviation = = = 3.5 Quartile Deviation = = = 3.5
Coefficient of Quartile Deviation= Coefficient of Quartile Deviation=
= = = 0.3684 = = = 0.35
Set B indicates less variability.
Merits and MERITS
DEMERITS

Demerits
 Not based on all the observations.
of Quartile  Easy to calculate and understand  Not representative value of data.
Deviation  Not affected by extreme values  It ignores the first 25% and last
as in the case of range. 25% of the observations.
 Very useful in estimating  Not a good measure of dispersion
dispersion in grouped data with since it depends on two position
open ended class measure.
 PRACTICE QUESTIONS

1. The water-tax bills (in

rupees) of 10 houses in a
locality are given below:
2. If first quartile = 37 and 3rd
144, 184, 130, 195, 132, 134, quartile = 69 Find (i) Quartile
196, 114, 212, 174 deviation (ii) Coefficient of
Find Quartile Deviation and quartile deviation.
Coefficient of Quartile
Deviation
3. Mean deviation

33
Mean Deviation
(Absolute Measure)

The mean deviation or the average deviation is defined as the mean of the absolute
deviations of observations from some suitable average which may be the arithmetic
mean or the median.
The difference (X- Average) is called deviations and when we ignore the negative sign,
this deviation is written as .
Mean deviation from mean=
M.D(about mean) =
Mean deviation from median=
M.D(about median) =
Mean
A relative measure of dispersion based on the mean deviation is
Coefficient
called theof
coefficient of the mean deviation or the mean coefficient
of dispersion. It is defined as the ratio of the mean deviation to the
Dispersion
average used in the calculation of the mean deviation.
(Relative
Mean Coefficient of Dispersion = =
Measure)
Median Coefficient of Dispersion==
Example 8: Sale of a coffee café in newYork city of 10 days are given as:
120,121,110,115,118,108,111,112,111,124 (in $).
(i) M.D(about mean) and Mean coefficient of dispersion
(ii) M.D(about median) and Median coefficient of dispersion.

X
108
Solution: (i) n = 10 110 5
=120,121,110,115,118,108,111,1 111 4
12,111,124 111 4
112 3
115 0
118 3
M.D(about mean) = 120 5
= 121 6
124 9
Total 46
 Example 8: Cont..

(ii) For median first we have to arrange the data:

X
108,110,111,111,112,115,118,120,121,124
108
Median = Value ofObservation
110 3.5
111 2.5
Median = Value ofObservation
111 2.5
= Value of 5.5th Observation
112 1.5
= 5th Obs +0.5(6thObs-5th Obs)
115 2.5
= 112 + 0.5(115-112)
118 4.5
= $113.5
120 6.5
M.D(about median) =
121 7.5
=
124 10.5
Total 47
Example 8: Cont..

Mean Coefficient of Dispersion =

= 0.04

Median Coefficient of Dispersion

==

Video Aid:
https://youtu.be/WpcocBR6Pr4
https://youtu.be/GdIkEngwGNU
  Shop 1
65,68,60,76,65,74,58, 56, 61,57
Example 9: The weight measurements in grams of 20 eggs are given. Compare the two group of eggs.
gms
= 1,4, 4, 12, 1, 10, 6, 8,3,  Shop 2
= 56 63,64,72,66,65,67,65,63, 67,67
M.D(about mean) = gms
= gms = 2.9,1.9,6.1.0.1,0.9,1.1.0.9,
Mean Coefficient of Dispersion = 2.9,1.1,1.1
= 0.0875 = 19
M.D(about mean) =
= gms
Mean Coefficient of Dispersion =
= 0.0288

The result shows the greater dispersion in weights of eggs in shop 1

Merits and Demerits of
mean Deviation DEMERITS
MERITS

 Mathematically illogical
 Easy to understand and to assume all negatives
compute. as positives.
 Based on all the observations.  Can’t be calculated with
 Is not affected by extreme distributions with open-
values. end class.
 Used to make healthy
comparisons
 1.
• Find the mean deviation from mean
and median for the following
values. Also find their coefficients.
• 30, 36, 32, 33, 35, 39, 36.5, 35 and
34

PRACTICE QUESTIONS • The weights in grams of 20 oranges

picked up at random from two
consignments are as follows
• Consignment 1: 126, 131, 113,

2.
82, 75, 204, 81, 84, 118, 104
• Consignment 2: 78, 90, 115, 110,
98, 106, 99, 107,84, 76
• Find (i) Mean Deviation (about
mean)
• (ii) Compare the variability in two
consignment with mean coefficient
of dispersion.
4. Standard deviation

42
The Standard
 Rather than relying on the location of the value, variance
Deviation and measures dispersion by calculating how far observations
Variance are from the mean.
(Absolute  Variance = Average of the distance from the mean of each
observation (squared).
Dispersion)  High variance means that many/most observations are far
from the mean but could be heavily influenced by outliers.
 Low variance means that many/most observations are
close to the mean.
The Standard deviation Takes square root of variance to put measure in the same
unit as the observations.
Example: The average rating of the Conservatives is 4.8
and the standard deviation is 2.8. This tells us that the
average amount that the ratings differ from the mean is 2.8
points on the 11 point scale used to measure feeling
towards the Conservative Party.
In contrast, the variance is 8.0, which can be interpreted as
8 squared points on the 11 point scale.
This explanation is confusing and has little intuitive power.

44
The Standard Deviation Standard Deviation of the observations is defined as the positive
square root of the mean of the squares of deviations of observations
taken from their arithmetic mean.
Cont…..
Let ‘X’ be a random variate which takes on ‘n’ values, the standard
deviation of these n observations is given by.

Where

45
 Alternatively,
The Standard Deviation

Cont. .. But the formula is used when the items are very small.
 The term is positive, therefore ‘S’ is always positive.
 If the original observations are in cm, the value of Standard
Deviation will also be in cm.
The Variance

The Variance of the observations is defined as the mean of the squares of

deviations of observations taken from their arithmetic mean.
In other words, the variance is the square of standard deviation and is
denoted by .
= Or Alternatively, =
If the original observations are in cm, the value of Standard Deviation will
also be in (cm)2
 Example 10 Company Market Value
(billions of dollars
PepsiCo 75
The following table gives the Google 107
2008 market values (rounded
to billions of dollars) of five PetroChina 271
international companies.
Find the variance and standard Johnson & Johnson 138
deviation for these data.
Intel 71

48
Solution

 Let x denote the 2008 market value (in billions of dollars) of a company.

X
=
75 5625
=
= 22920 107 11449

= 22920=5390.24 271 73441

to obtain the standard deviation 138 19044

71 5041
 Thus, the standard deviation of the market values of
these five companies is $82.08 billion.

49
Example 11:
Following are the 2009 earnings (in thousands of dollars) before taxes for all six employees of a small
company.
88.50 108.40 65.50 52.50 79.80 54.60
Calculate the variance and standard deviation for these data.

X
Solution:
88.50 13.62 185.5
108.40 33.52 1123.59
65.50 -9.38 87.98
= = =388.9
52.50 -22.38 500.86
thousand
79.80 4.92 24.20
=$19,721 54.60 -20.28 411.27
 Thus, the standard deviation of the 2009 earnings =2333.4
of all six employees of this company is
 $19,721.
50
Practice Questions
1.
• Find the sample variance and standard deviation for the
amount of European auto sales for a sample of 6 years
shown. The data are in millions of dollars. 11.2, 11.9,
12.0, 12.8, 13.4, 14.3

2.
• The following data are the ages (in years) of six students.
• 19 19 19 19 19 19 Calculate the standard deviation. Is its
value zero? If yes, why?

3.
• The following data give the number of highway
collisions with large wild animals, such as deer or
moose, in one of the northeastern states during each
week of a 9-week period.
• 7 1 03 8 2 5 7 4 9
• Find the variance,and standard deviation.
 Coefficient of Variation (Relative Measure)

 The most important of all the relative measures of dispersion is the

coefficient of variation.
 Coefficient of Variation is a measure of spread that describes the amount
of variability relative to the mean.
 The coefficient of variation (C.V) is defined as:
 C.V = *100 =
 It is important to note that the coefficient of variation is always a
percentage.
 The coefficient of variation has a great practical significance and is the
best measure of comparing the variability of the two series.
Uses of
Coefficient of
Variation  The series or groups for which the coefficient of variation
is greater is said to be more variable (or less consistent).
 On the other hand, the series for which the variation is
less is said to be less variable (or more consistent).
 The C.V is used to compare the dispersion in different sets
of data particularly the data which differ in their means or
differ in the units of measurement.
• For example , suppose an
automobile dealer wanted to
compare the standard deviation
of miles driven for the cars she
received as trade-ins on new cars.
She found that for a specific year,
the standard deviation for
Mercedes S-Class was 422 miles
and the standard deviation for
Audi A8. was 350 miles. She
could say that the variation in
mileage was greater in the
Mercedes S-Class.
• But what if a manager wanted to compare the standard deviations of two
different variables, such as the number of sales per sales- person over a
3-month period and the commissions made by these salespeople?
• A statistic that allows you to compare standard deviations when the units
are different, as in this example, is called the coefficient of variation.
Example 12: Ten measurements gave the following results:
Length in cm: 77,73,75,70,72,76,75,72,74,76
(i) Find Variance and Standard Deviation
(ii) Find Coefficient of Variation
X
Solution:
First we have to calculate 77 3 9
73 -1 1
75 1 1
70 -4 16
72 -2 4
(i) = = = 4.4 (Cm)2 76 2 4
75 1 1

= 2.0976 cm
72 -2 4
74 0 0
76 2 4
(ii) C.V = = = 2.8345 %
Total 44
Example 13: The quarterly profits (in lakhs of Rs.) of two companies are given below:
Calculate their Standard Deviation and compare the variability of two companies.
Quarters Company A Company B
Jan-March 9.5 15.5
April-June 13.7 21.2
July-September 10.4 23.4
October-December 8.6 18.8

=9.5+13.7+10.4+8.6=42.2 =15.5+21.2+23.4+18.8=78.9
=(9.5)2+(13.7)2+(10.4)2+(8.6)2=460.06 =(15.5)2+(21.2)2+(23.4)2+(18.8)2=397.6275

= = = =
= Rs. 1.9267 = Rs. 2.9321
C.V (X) = = = 18.26 % C.V(Y) = = = 14.86 %

By comparing the coefficient of variations, it can be observed that Company B provides the more
Consistent Performance.
Example 14: A company has two sections with 40 and 65 employees respectively. Their average weekly wages are $450 and
$350. The standard deviation are 7 and 9.
(i) Which section has a larger wage bill?
(ii) Which section appears to be more consistent in performance?

Section A Section B
(i) Wage Bill section A (i) Wage Bill section B
= =
450 x 40 = $18000 350 x 65 = $22750
(ii) Variability of A (ii) Variability of B
C.V= C.V=
C.V = C.V=
C.V = 1.56% C.V= 2.57%
Section A appears to be more consistent in Section B has larger wage bill
performance as having less value of C.V.
Video Aid:
https://www.youtube.com/watch?v=wDAd_QHKoOg
Merits and MERITS DEMER I TS

Demerits of Difficult to compute unlike

Standard Takes all data into account. other measures of dispersion.
Has greater mathematical
Deviation significance and capable of further
Hard to interpret.
mathematical treatments. Give more weightage to
Least affected by fluctuation of extreme values.
sampling.
Not expressed in squared units, so
makes more sense descriptively.
 PRACTICE QUESTIONS
• A random sample of 101 weekly personal
incomes in city A and city B gives the following

1.
results:
• City A: mean =100, S=30
• City B: mean = 79.93, S=13.49 and median=75
• In which city is the distribution of income
more variable?

• The following data shows the temperature

in (degrees Celsius) recorded in an

2.
experiment in the laboratory:
81,94,64,80,75,69,96,66,80,91,85,79
Find, Variance, Standard Deviation and
Coefficient of Variation
 (i) The standard deviation is zero if all the observations has some
constant value. If ‘C’ is a constant, then
S.D(C)=0
Properties of (ii) The standard deviation do not change by change of Origin
Standard Deviation S.D(X A) = S.D(X), where A is a constant & S.D(A) =0
S.D(X - A) = S.D(X)
(iii) The standard deviation is affected by change of Scale
S.D(AX) = S.D(X), where A is a constant.
Also S.D() = S.D(X)
(iv) If X and Y are independent random variables then,
S.D (X Y) = S.D(X) + S.(Y) ,
where a and b are constants.
Properties of (V) If we have two sets of data with n1 and n2 observations having
means and with variances and respectively, then the
Standard combined standard deviation, SC of all the n = n1 + n2
Deviation observations is given by:

Cont…..
Alternatively,

Where =
 

Video Aid:
https://youtu.be/ctEGvROA7Bg
https://youtu.be/JYKcTA_IL1I
Example 5:

The mean and standard Group Mean Standard Deviation

deviation of the marks 1 60 8
obtained by two groups
of students, consisting of 2 55 7
50 each, are given below: = = = 57.5
Calculate the mean and
standard deviation of the
=
marks obtained by all the
100 students =