MAT 161 - DESCRIPTIVE STATISTICS

LECTURE 4: MEASURES OF DISPERSION

LECTURE C24-1-4 :: WED., FEB. 21, 2024 11:00 AM (WAT) - WED., FEB. 21, 2024
7:00 PM (WAT)

Prepared by Dr. Akinwale Lewis OLUTIMO

Email: akinwale.olutimo@lasu.edu.ng | Phone: 08139718341
Mathematics Department, Lagos State University.

0:10
NOTE: Fractiles
The median is the middle value of an ordered data. Associated with median
are the quartiles. The fractiles are the quartiles, deciles and percentiles which
are measures that divide a distribution into specified fraction of the
distribution. These measures are also known as fractiles.
MODE
The mode of set values is that value which occurs most frequently. It is a
measure of central tendency.

0:10
Example 1
Find the mode of: 1, 5, 5, 2, 5, 3, 4, 7, 5, 6, 5, 8, 4, 2, 2
Solution:
The mode is 5 since the value 5 has the largest frequency.

Example 2: Find the mode of the data given in table below.

x 1 2 3 4 5 6 7
f 2 10 18 32 4 1 3

0:10
 Solution:
4 is the mode since the value has the largest frequency of 32. This simply
means that out of total occurrence of 70, the 4 occurred 32 times.

The Mode for Grouped Data

The modal class of a grouped frequency distribution is the class with the
greatest frequency. The value of the mode within the modal class may be
estimated from the histogram of the distribution as shown in the diagram
below.

0:10
0:10
 To calculate the value of this mode, use:

△1 △1
M ode = Lmode + ×c M ode = Lmode + ×c
△1 +△2 △ 1 +△ 2

where: Lmode
Lmode = Lower class boundary of modal class

△ △1 = Frequency of modal class - Frequency of class preceding the modal

1

class
△ △2 = Frequency of modal class - Frequency of class succeeding the modal
2

class
c = Width of modal class

0:10
 Estimate the mode for the following data

Weight (Kg) 10-15 15-20 20-25 25-30 30-35 35-40

Frequency 7 12 28 49 38 11
Lmode Lmode= 24.5
△ △1 = 49-28
1

△ △2 = 49-38
2

c=5

0:10
 21 21
M ode = 24.5 + × 5 M ode = 24.5 + × 5
21+11 21+11

= 24.5+3.28
=27.78 kg
Relationship between Mode, Median and Mean
For a normal distribution, the mean median and mode are all the same. The
median is the middle items and always be in the middle of mean and mode as
shown below. When the distribution is normal, the mean, median and mode
will overlap with an insignificant different from another. The relationship
between mode, median and mean can be expressed as follows:

0:10
0:10
0:10
 Measures of Dispersion (Scale or Variation)

It is not enough to summarize data by their central tendencies alone, it is

equally important to know the spread in the data.

Some of the measures of scale (dispersion or variation) include: Range; Mean

Deviation, Variance, Standard Deviation, Quartile Deviation, and so on.
Dispersion is the extent to which a set of data is spread-out, while Variation is
the extent to which a set of data spread around its mean or average value.

0:10
Measures of Scale are summary values for determining numerically the extent
of this spread. The larger these values are, the larger the spread in the data,
and by extension, the more the level of uncertainty in the data. The smallest
value of these measures is zero(0), where in this case, the data values do not
vary.

0:10
 RANGE (R)
The range of a set of data is the difference between the highest value of data
(H) and the lowest value of data (L) in the data-set. That is, Range(R)=H-L.
Example
Find the range of the following set of data: 23, 20, 18, 24, 27, 34, 30 and 24.
Solution
The highest value of data, H=34
The lowest value of data, L=18
R= H- L= 34-18= 16.

0:10
 Mean Deviation (M.D)
As the name implies, the mean deviation of a set of data is the mean of the
absolute deviations of the data from their arithmetic mean. Thus, for data-set
x1 , x2 . . . xn , the mean deviation (M.D) is given by
¯
Σ|xi −X |
M. D =
n

Mean Deviation from Frequency Distribution

¯
Σf|x−X |
M. D =
Σf

0:10
 Example
Find the mean deviation of the data 23, 20, 18, 24, 27, 34, 30, 24
Solution:
23+20+18+24+27+34+30+24
First, we find the mean as X
¯ 200
= = = 25
8 8

x 23 20 18 24 27 34 30 24 Σ

¯
xi − X -2 -5 -7 -1 2 9 5 -1 0

|x − X̄ | 2 5 7 1 2 9 5 1 32

0:10
 ¯
Σ|xi −X |
M. D =
n

.
32
M. D = = 4
8

Example
The distribution of the weights of some students in a private University is
given in Table 4.1 . Calculate the mean deviation for this data.
Table 4.1

Weight(Kg) 60-62 63-65 66-68 69-71 72-74

Frequency 15 28 27 22 8

0:10
Weight(kg) Mid-Point(x) f fx x − X̄ |x − X̄ | f |x − X̄ |

60-62 61 15 915 -5.4 5.4 81.0

63-65 64 28 1795 -2.4 2.4 67.2
66-68 67 27 1809 0.6 0.6 16.2
69-71 70 22 1540 3.6 3.6 79.2
72-74 73 8 584 6.6 6.6 52.8
Σ 100 6640 296.4

0:10
 The mean is X̄ kg
6640
= = 66.40
100
¯
Σf|x−X |
and M . D =
Σf

kg
296.4
M. D = = 2.964
100

Variance:

The Variance of a set of data is the arithmetic mean of the squared deviations
from the arithmetic mean X̄ of the set. Thus,
¯ 2
Σ(xi −X )

where x1 , x2 . . . xn are the data-items in the set.

0:10
 Standard Deviation (SD)
This is the positive square-root of the value of variance. That is,
−−−−−−−
S. D = √variance

NOTE
The unit of Variance is the square of the unit of data, while the unit of
Standard deviation is the unit of data.
Example
Determine the variance and standard deviation of following set of data: 23, 20,
18, 24, 27, 34, 30 and 24 which has X̄ = 25 from previous example.

0:10
 Solution
x 23 20 18 24 27 34 30 24 Σ

¯ 2
(xi − X ) 2 25 49 1 4 81 25 1 190
¯ 2
Σ(xi −X )
Variance= n
190

8
= 23.75 square units
−−−−
S. D = √23.75 = 4.8 units

0:10
 Variance and Standard Deviation of Grouped Data
2
2
Σfx Σfx
V ariance =
Σf
− (
Σf
) and
−−−−−−−
S. D = √variance .
Example
Find the variance and Standard deviation of the data set in Table 4.1 above

0:10
Solution:
Weight(Kg) Mid-Point (x) f fx fx
2

60-62 61 15 915 55815

63-65 64 28 1792 114688
66-68 67 27 1809 121203
69-71 70 22 1540 107800
72-74 73 8 584 42632
Σ 100 6640 442138

0:10
 2
2
Σfx Σfx
V ariance = − ( )
Σf Σf

442138 6640
V ariance = − ( )
100 100

= 12.42 kg
2
and
−−−−−−−
S. D = √variance
−−−−
S. D = √12.42 = 3.52 kg.
NOTE:
M = Q2 = D5 = P50

0:10
 Quartile and Percentile Deviations
These are particularly useful for determining the spread in a data-set which is
perceived to have been marred by extreme values or outliers.
They include:

Interquartile Range (IQR) : I QR = Q3 − Q1

Semi Interquartile Range (SIR): SI R = 1

2
(Q3 − Q1 )

Percentile Deviation (PD): P90 − P10

0:10
 Relative Measures of Dispersion
This is used to compare the spread in the data of samples measured in
different units. It is usually in the form,

Spread
Coefficient of Spread= Centre

Coefficient of Mean Deviation (CMD): =CM D =

M .D

¯
X

Q3 −Q1
Coefficient of Quartile Deviation (CQD): CoD = =
SIR

¯
Q +Q X
3 1

0:10
 Coefficient of Standard Deviation (CSD):
SD
CSD =
¯
X

Coefficient of Variation (CV): CV

SD
= × 100
¯
X

LASU-VLAP
Your attendance for this lecture can not be marked at this moment because it was scheduled for Wed., Feb. 21,
2024 11:00 AM (WAT) - Wed., Feb. 21, 2024 7:00 PM (WAT) and you are outside the specified lecture period.

OK

0:10