Our Lady of Fatima University

GRADUATE SCHOOL
Department of Education and Teaching
MABC 202 – (Dispersion)

Measures of Variability
A measures of variability, also called measures of dispersion, It is a statistic calculated from a
set of observations or scores and designed to show the extent to which the individual observations or
scores are concentrated about or scattered from the mean or other measures of central tendency.
A group whose variability is smaller is more homogeneous than a group with a bigger variability.

Range - The difference between the highest and the lowest score in a set of ungrouped data.

Example. Given the following sets of data, 23, 18, 24, 17, 29, 19, 15, 20, compute for the value of
Range.
Solution:
R = HS – LS = 29 – 15 = 14

Interquartile range is the distance between the quartiles Q1 and Q3. The semi-interquartile range
called quartile deviation which is more commonly used as a measures of variability, also includes
Decile Deviation and Percentile deviation.

Interquartile Range Quartile Deviation

Q − Q1
IR = Q3 – Q1 QD = SIR = 3
2
Decile Deviation (DD) Percentile Deviation (PD)
DD = D9 – D1 PD = P90 – P10

Example. Given the following sets of data, 23, 18, 24, 17, 29, 19, 15, 20, compute for the value of
Quartile Deviation
Solution:
15, 17, 18, 19, 20, 23, 24, 29
1 3 Q − Q1 23 − 17
Q1 = 8 = 2nd; Q1 = 17 * Q3 = 8 = 6th; Q3 = 23 * QD = 3 = =3
4 4 2 2

Variance – It is a reliable measure of dispersion and is widely used in modern educational statistics
and educational experiments.
UNGROUPED DATA
Population Variance (  2) Sample Variance (s2)

 =2  (x − x)2
s =
2  (x − x)2
N n −1
Remark: Population Standard Deviation 𝝈= √𝛔𝟐 & Sample Standard Deviation, s = √𝒔𝟐

𝜎=√  𝑠=√
(x − x) 2 (x − x) 2
N n −1
Example. Given the following sets of data, 23, 18, 24, 17, 29, 19, 15, 20, compute for the standard
deviation (s).

1|Page
Solution:
n

x i
23 + 18 + 24 + 17 + 29 + 19 + 15 + 20 165
x= i =1
= = = 20.63 *
n 8 8

s=
 ( x − x) 2

=
(2.37) 2 + (−2.63) 2 + (3.37) 2 + (−3.63) 2 + (8.37) 2 + (−1.63) 2 + (−5.63) 2 + (−0.63) 2
n −1 7

141.8752
= = 20.26788571 = 4.50 s tan dard deviation
7

Note that the Variance is (s2) = 20.27

Coefficient of Variation – The coefficient of variation is a measure of spread that describes the amount
of variability relative to the mean. Because the coefficient of variation is unitless, you can use it instead
of the standard deviation to compare the spread of data sets that have different units or different means.

It is the ratio of the standard deviation to the absolute value of the arithmetic mean, expressed as
percentage. It is use to compare the relative variation of even unrelated quantities.

 s 
c.v. =  x100  %
/ x / 

For example, you are the quality control inspector at a milk bottling plant that bottles small and large
containers of milk. You take a sample of each product and observe that the mean volume of the small
containers is 1 cup with a standard deviation of 0.08 cups, and the mean volume of the large containers
is 1 gallon (16 cups) with a standard deviation of 0.4 cups. Although the standard deviation of the gallon
container is five times greater that the standard deviation of the small container, their coefficients of
variation (CVs) support a different conclusion:

Large container Small container

CV = (0.4/16)*100 CV = (0.08/1)*100
CV = 2.5% CV = 8%

The coefficient of variation of the small container is more than three times greater than that of the large
container. In other words, although the large container has a greater standard deviation, the small
container has much more variability relative to its mean.

Example. Given the following sets of data, 23, 18, 24, 17, 29, 19, 15, 20, compute for the value of
coefficient of variation.
Solution: Answer. cv = 21.81%

2|Page
Skewness – For a symmetrical distribution, the mean, median, and mode are located at exactly the
same point in the distribution, thus when graph it shows a symmetrical bell shape graph. Some
distributions are asymmetrical or skewed. Graph may be skewed to the right or left, moreover
asymmetrical distribution will tell us that the mean, median, and mode are at different points.

Mean Mo Mdn Mean Mean Mdn Mo

Median POSITIVELY SKEWED NEGATIVELY SKEWED
Mode
Symmetrical Asymmetrical
(-0.49<SK<0.49)

Skewness can be measured using the formula:

 (x − x )
3
n n −1
SK = •
 (x − x) 
;
n−2 3
2 2

if the SK is positive then it is positively skewed or skewed to the right, if SK is negative it is

negatively skewed or skewed to the left, and if SK is 0 then the distribution is approximately
symmetrical.

Example. Compute the Skewness of the following scores: 2, 3, 4, 3, 9.

Solutions:
x (𝑥 − 𝑥̅ )2 (𝑥 − 𝑥̅ )3
x=
 x = 2 + 3 + 4 + 3 + 9 = 21 = 4.2 * 2 4.84 -10.648
n 5 5
3 1.44 -1.728
4 0.04 -0.008
3 1.44 -1.728
9 23.04 110.592
Sum 30.8 96.48

 (x − x )
3
n n −1 5 4 96.48
SK = • = • = 3.333 • 0.5644 = 1.88 ;
n−2
 (x − x)  30.82
3 3
2 2 3

Hence it is positively skewed.

Kurtosis – This refers to the peakedness exhibited by the frequency distribution.

Leptokurtic (ku > 3.50)
3|Page
 Mesokurtic (2.50 < ku < 3.49)

Platykurtic (ku < 2.49)

The peakedness can be measure using the formula:

Ku =
n(n + 1)
•
 ( x − x) − 3(n − 1) 2
4

(n − 1)(n − 2)( n − 3) s4 (n − 2)( n − 3)

An approximately mesokurtic distribution has a kurtosis of 3 (2.50<Ku<3.49). Thus, a kurtosis

of more than 3 (ku > 3.50) means that the distribution is leptokurtic, and kurtosis of less than 3 (Ku <
2.49) means that the distribution is platykurtic.

Example.8. Compute the kurtosis of the following scores: 2, 3, 4, 3, 9.

x ̅) 2
(x-𝒙 ̅ )4
(x-𝒙
2 4.84 23.4256
3 1.44 2.0736
4 0.04 0.0016
3 1.44 2.0736
9 23.04 530.8416
s2=∑(x − 𝑥̅ )2
(n-1)
= 30.8
4
2
s = 7.7

s4= ( S2)2
̅=
𝒙 s4 = (7.7)2 ̅) 𝟒 =
∑(𝐱 − 𝒙
4.2 s4 = 59.29 558.416

Ku =
n(n + 1)
•
 ( x − x) − 3(n − 1) 2
4

(n − 1)( n − 2)( n − 3) s4 (n − 2)(n − 3)

5(6) 558 .416 3(4) 2

Ku = • − = 1.25 • 9.418 − 8 = 3.77 , hence it is leptokurtic
(4)(3)( 2) 59.29 (3)( 2)

Prepared by: Christopher J. Delino, Ph.D.

Leading you to the top.

4|Page