MT.

PLEASANT HIGH SCHOOL

A Level Geography - Geostatistics Seminar
Enhancing student performance in A Level Geography
(6037/3).
Assessment Specialist: Mr. N Karimazondo -0774114549
Measures of Variability
Measures of Dispersion (Variability)

 Measures of dispersion are also called measures of variability, measures of

spread or measures of scatter. Measures of dispersion describe the
scatterdness of data.
 The measures of variability include the range, the variance, the standard
deviation and the coefficient of variation.

The range
 The range is simply the difference between the highest score and the lowest
score in a data set.
 The problem of the range is that it is likely to give misleading results on many
occasions because the extreme values are frequently eccentric and atypical of
the rest of the distribution.
Advantages of the range
Advantages of the range are that:
 All the same, the range is the easiest, among all the measures of variability to
compute
 It provides a rough idea of the variability using the scores from either end of the
distribution.

Disadvantages of the range

Disadvantages of the range are that:
• It is influenced by extreme scores.

• It might give a misleading picture.

• It is a poor measure of variability since it only considers the largest and the least
scores.
• It is a crude measure that should not generally be relied upon in applied social
sciences.
Examples

1. The range of 65,73,89,56,73,52,47 is 89-47 = 42.

2. If the highest score in a 1st year statistics exam was 98 and the lowest 48, then
the range would be 98-48 = 50.
3. The range of the set of numbers; 8, 4, 10, 12, 6, 7, 8, 15, 14, is 15 – 4 = 11.

From the above examples, we can see that the greater the range the greater the dispersion.
Range is easy to understand, and its computation is simple, but is not based on all
observations. Also, range doesn’t change if values of a set of data change provided the two
extreme values remain the same.
The variance
 Variability refers to the difference between each score and every other score in
the distribution (data set).
 Calculating variability using this definition would be very tedious though. Instead,
since the mean is a summary of the collective location of the scores, we consider
variability in terms of the deviation of each score from the mean.
 That is, in terms of how each score differs from the mean.
 Squaring the deviations and finding their average provide an excellent measure of
variability in form of what we term variance.
 Variance can, therefore, be defined as the average of squared deviations from the
mean.
 The advantages and disadvantages of the variance are similar to those of the mean
since its definitional computation largely relies on the mean.
Variance for ungrouped data.
For n observations x1, x2, x3, ... , xn the sample variance is given by the formulae

Variance (s2) = Σ (x i ̶ x)2

n ̶ 1

Example

Find the variance of the set of numbers; 8, 4, 10, 12, 6, 7, 8, 15, 14,
xi
(x i ̶ x ) (x i ̶ x )2
8 − 1.33 1.78
4 − 5.33 28.41
10 0.67 0.45
12 2.67 7.13
6 − 3.33 11.09
7 − 2.33 5.44
8 − 1.33 1.78
 15 5.67 32.15
14 4.67 21.78
Σ = 84 Σ =110.01

x = Σxi
n

x = 84
9

= 9.33
Variance (s2) = Σ (x i ̶ x)2
n ̶ 1

𝟏𝟏𝟎.𝟎𝟏
= 𝟖.𝟑𝟑

= 13.21
3.6.2 Variance for Grouped data.
For n class intervals with class centre or mark x1, x2, x3, ... , xn and the corresponding
frequency f1, f2, f3, ... , fn the sample variance is given by the formulae

Variance (s2) = Σ fi (x i ̶ x)2

n ̶ 1

where n is the total frequency

Example
A case of 300 oranges was sampled, opened and each orange was weighed. The weight of
each orange was measured to the nearest gram and recorded in the following table.
Weight No. of Oranges
(grammes)
60-79 20
80-99 60
100-119 80
 120-139 50
140-159 60
160-179 30
.

Calculate the variance of the distribution.

Solution
Frequency Distribution
Weight No. of Class fixi
(grammes) Oranges f i Centre( x i) (x i ̶ x) (x i ̶ x )2 fi(x i ̶ x
)2
59.5-79.5 20 69.5 1 390 − 50.7 2 570.5 51 410
79.5 -99.5 60 89.5 5 370 − 30.7 942.5 56 550
99.5-119.5 80 109.5 8 760 − 10.7 114.5 9 160
119.5-139.5 50 129.5 6 475 9.3 86.5 4 325
139.5-159.5 60 149.5 8 970 29.3 858.5 51 510
159.5-179.5 30 169.5 5085 49.3 2430.5 72 915
Total 300 36050 245 870
x = Σxi
n

x =

= 120.17

Variance (s2) = Σ fi (x i ̶ x)2

n ̶ 1

= 245870
299

= 822.31
The standard deviation
To achieve a measure of variability which is in the same units as the original data, we
have to unsquare the variance. This is done by obtaining the square root of the variance.
The resultant positive square root is what is termed the standard deviation. The standard
deviation is, therefore, the appropriate average of deviations of the scores from the
mean. Its advantages and disadvantages are also akin to those of the mean.

𝑠𝑑 = √𝑉𝑎𝑟𝑖𝑎𝑛𝑐𝑒

= √822.31

= 26.68
5.4.4 Coefficient of variation
The coefficient of variation is the ratio of the standard deviation to the mean. It is
expressed as a percentage and it is used for comparing variations between sets of
data or between samples from the same population.

𝑠𝑑
𝑐𝑣 = × 100
𝑥̅

26.68
= × 100
120.17

= 22.20%

Activity
1. In your own words, explain the following descriptive
concepts:
(a) Measures of variability
(b) Variability
 (c) Variance
(d) Standard deviation
(e) Coefficient of variation
(f) Range
2. Outline the advantages and disadvantages of the following
descriptive statistics: a) Range
b) Variance
c) Standard deviation