UNIVERSITY OF ILESA, ILESA, OSUN STATE, NIGERIA

FACULTY OF COMPUTING
DEPARTMENT OF CYBERSECURITY
DEPARTMENT OF ICT
DEPARTMENT OF SOFTWARE ENGINEERING

STA 111: Descriptive Statistics (3 Units C: LH 45)

4.0. MEASURES OF DISPERSION (OR SCATTER OR VARIABILITY)

Introduction
Measures of dispersion describe how spread out or scattered a set or distribution of numeric data
is.
Measures of dispersion include the range, quartile deviation, mean deviation, variance, standard
deviation and coefficient of variation.
Range
The range is the difference between the highest and the lowest score in a given data.
Example
In STA 111 class, the following are the scores of sampled ten students in a test:
29, 18, 58, 62, 75, 75, 35, 40, 40, 65.
The range is 75 – 18 = 57
&% and 18 are the highest and the lowest scores respectively and are the extreme values. Range
is not a reliable measure of dispersion because it uses only the extreme values of a distribution.
Quartile Deviation
𝑄1 −𝑄2
This is also called semi-interquartile range. It is defined as , where 𝑄1 𝑎𝑛𝑑 𝑄2 are the
2
lower and upper quartiles respectively. It gives an idea of how much the central values of the
data deviate from the median.
Example Calculation
Consider the dataset:
5,7,8,12,15,18,20,24,28, 30
Arranged data: Already in ascending order.
1. 𝑸𝟏 (25th percentile): 𝑸𝟏 = 8
2. 𝑸𝟑 (75th percentile): 𝑸𝟑 =24
3. Quartile Deviation: QD= 24 – 8 = 16
Mean Deviation (Now known as Mean Absolute Deviation)
This is the mean dispersion that gives the average absolute difference between each item and the
mean. It measures the average distance between each data point and the central value (mean or
median) of the dataset. It shows how much the values in a data set deviate from the central
tendency, giving an idea of the spread or dispersion of the data.
∑ 𝑓|𝑥−𝑥̅ |
M.D. = ∑𝑓

Mean deviation attempts to find the spread about the mean. It is expected that data which spread
closely the mean would have small dispersion and those with values spread around the mean will
have large dispersion.
Example
Find the mean deviation of the following numbers:
7.2, 10.7, 8.5, 6.2, 9.8, 7.4, 5.7, 4.6, 6.5, 7.7 and 11.5
Solution
85.8
𝑥̅ = = 7.8
11
x 7.2 10.7 8.5 6.2 9.8 7.4 5.7 4.6 6.5 7.7 11.5
|𝑥 − 𝑥̅ | 0.6 2.9 0.7 1.6 2.0 0.4 2.1 3.2 1.3 0.1 3.7

M.D. ∑|𝑥 − 𝑥̅ | = 18.6

∑|𝑥−𝑥̅ | 18.6
M.D. = =
𝑁 11

= 1.69
Sometimes the mean deviation can be calculated from the median. The method of calculation is
the same except that the median is used instead of the arithmetic mean.
Variance and Standard Deviation
The standard deviation is the most important of all measures of dispersion. Equally important is
the variance which is the square of the standard deviation. The variance is the squared deviation
of the individual value from the mean and is denoted by S2 when we consider only a sample of a
group (sample less or equal to 30) or with 𝜎 (𝑠𝑖𝑔𝑚𝑎) when we study entire group.
 1
𝜎 2 = 𝑁 ∑(𝑥 − 𝜇)2 , 𝑤ℎ𝑒𝑟𝑒 𝜇 is the population mean

When n is small (≤ 30), S2 is normally taken as the variance and is obtained by:
1
𝑆 2 = 𝑁−1 ∑(𝑥 − 𝑥̅ )2

In general, when the population mean 𝜇 is not known, the sample mean 𝑥̅ is used as the estimate.
The variance measures the extent to which he daa differ among themselves from the mean. If 𝑋𝑖 𝑠
are all identical, there can be no differences and the estimate of the variance 𝑆 2 = 0 If they differ
only slightly small, the variance is small and if widely, the variance is large.
Standard deviation is the root mean square deviation and is denoted by

(𝑥−𝑥̅ )2
𝑆=√ 𝑁−1

For a large population,

(𝑥−𝑥̅ )2
𝜎=√ 𝑁

For a frequency distribution, the variance and the standard deviation are given as

∑ 𝑓(𝑥−𝑥̅ )2 ∑ 𝑓(𝑥−𝑥̅ )2
𝜎2 = ∑𝑓
and 𝜎 = √ ∑𝑓
respectively.

It can be shown from the above that

∑ 𝑓𝑥 2 ∑ 𝑓𝑥 2
𝜎2 = ∑𝑓
− ( ∑𝑓 )

∑ 𝑓𝑥 2 ∑ 𝑓𝑥 2
And standard deviation = √ ∑𝑓
− ( ∑𝑓 )

Example
Compute the variance and standard deviation for the following test scores:
Scores 10 -19 20 – 29 30 – 39 40 – 49 50 – 59 60 – 69 70 - 79
Frequency 2 4 4 8 6 3 2

Solution
Class f Class mark fx 𝒙−𝒙 ̅ ̅) 𝟐
(𝒙 − 𝒙 ̅) 𝟐
f(𝒙 − 𝒙
10 – 19 2 14.5 29 -30 900 1800
20 – 29 4 24.5 98 -20 400 1600
30 – 39 4 34.5 138 -10 100 400
40 – 49 8 44.5 356 0 0 0
50 – 59 6 54.5 327 10 100 600
 60 – 69 3 64.5 193.5 20 400 1200
70 - 79 2 74.5 149 30 900 1800
29 1290.5 7400

∑ 𝑓𝑥 1290.5
𝑥̅ = = = 44.5
∑𝑓 29
∑ 𝑓(𝑥 − 𝑥̅ )2 7400
𝜎2 = = = 255.17
∑𝑓 29

S.D. = (𝜎) = √255.17 = 15.97

Calculation of Variance using Assumed mean Method
If A is the assumed mean and 𝑑 = 𝑥 − 𝐴, then
∑ 𝑓𝑑2 ∑ 𝑓𝑑 2
𝜎2 = ∑𝑓
− ( ∑𝑓 )

Using coding method, we have the formula as

𝑥−𝐴
𝜎𝑥 2 = 𝑐 2 𝜎𝑢 2 where U= 𝑐

Example
Find the (i) Variance (ii) standard deviation of the following data using assumed mean method
Score 50 -54 55 – 59 60 – 64 65 – 69 70 -74 75 – 79 80 -84 85 – 89 90 - 94
Frequency 3 5 8 10 7 6 3 3 1

Solution
Let the assumed mean = 72
Score f Class mark d = X -A fd fd2
X
50 -54 3 52 -20 -60 1200
55 – 59 5 57 -15 -75 1125
60 – 64 8 62 -10 -80 800
65 – 69 10 67 -5 -50 250
70 – 74 7 72 0 0 0
75 – 79 6 77 5 30 150
80 – 84 3 82 10 30 300
85 – 89 3 87 15 45 675
90 – 94 1 93 20 20 400
46 -140 4900
 ∑ 𝑓𝑑2 ∑ 𝑓𝑑 2
𝜎2 = ∑𝑓
− ( ∑𝑓 )

4900 −140 2
= −( )
46 46

= 106.52 – 9.263
= 92.257

S.D. = (𝜎) = √92.257 = 9.861

Using coding method:
A = 72, c = 5
Score f Class mark 𝒖=
𝑿−𝑨 fu fu2
𝒄
X
50 -54 3 52 -4 -12 48
55 – 59 5 57 -3 -15 45
60 – 64 8 62 -2 -16 32
65 – 69 10 67 -1 -10 10
70 – 74 7 72 0 0 0
75 – 79 6 77 1 6 6
80 – 84 3 82 2 6 12
85 – 89 3 87 3 9 27
90 – 94 1 93 4 4 16
46 -28 196

𝜎𝑥 2 = 𝑐 2 𝜎𝑢 2
∑ 𝑓𝑢2 ∑ 𝑓𝑢 2
= 𝑐2 ( ∑𝑓
− ( ∑𝑓 ) )

196 −28 2
= 52 ( 46 − ( 46 ) )

= 25(3.89)
= 97.25
Coefficient of Dispersion )or variation)
𝜎
Coefficient of dispersion (c.d) = 𝑋
𝜎
Coefficient of variation (c,v) = 𝑋 × 100%

Example
In the above example, find the coefficient of dispersion and the coefficient of variation for class
75 - 79
Solution
𝜎 = 9.86
For the class 75 -79, X = 77
𝜎
Hence, c.d = 𝑋
9.86
= 77

= 0.128
c.v. = 0.128 × 100%
= 12.8%
Exercise
1. Calculate the
i. Mean deviation
ii. Variance
iii. Standard deviation
for the following data:
No of Sales 0–4 5–9 10 -14 15 -19 20 – 24 25 - 29
No of Salesmen 1 14 23 21 15 6

2. Using coding method, calculate the variance and standard deviation for the following
data:
Class 10 – 14 15 – 19 20 -24 25 -29 30 – 34 35 - 39
Frequency 4 9 18 17 12 3