MEASURES OF DISPERSION

It can be seen from the previous lesson that it is useful to locate the center of a set of data
to be able to describe it. You must not be contented with identifying the central point of a
distribution. Interpretation of the data can also be enhanced if you know how to spread the
individual observations around that center point.

In this lesson, I will discuss another type of measure, which will enable you to know more
about the shape or the spread of the distribution (whether the group is homogeneous or
heterogeneous in terms of some aspect). These measures are called measures of dispersion.

A measure of dispersion indicates the degree of spread of a given set of

data.

Let us examine three small sets of data.

Data Set 1 Data Set 2 Data Set 3

0 3 8
2 5 4
13 7 3

You can see that the mean for each data set is 5. Can you then conclude that the data are
similar? Of course not. However, when you are only told of the means without looking at the
individual observations, you might presume a similarity.

For you to have a more complete idea of the data sets, you need a measure of how spread
out the observations are from the mean of 5.

The measure of dispersion indicates to what degree the individual observations are
dispersed or spread out around their mean. They identify whether the distribution is homogeneous
or heterogeneous. From the three data sets, Data Set 1 has a bigger measure of dispersion than
Data Set 2 and 3 became the observations.

In this lesson, I will discuss four measures of dispersion, namely: range, mean deviation,
standard deviation, and variance.

RANGE

The simplest measure of dispersion is the range. The range is the difference between the
highest and the lowest values in a set of observations.

There are two types of range depending on whether the data are discrete or continuous.

1. Exclusive Range (Re) - it is the difference between the highest value (H) and the lowest
value (L).
Re = H - L
 Example: Find the exclusive range for the Chemistry and Biology scores given below:

Chemistry Scores: 7, 9, 11, 8, 13, 19, 15, 5, 8, 10

Biology Scores: 10, 14, 14, 11, 15, 16, 16, 14, 12, 13

Solution:
Re (Chemistry) = 19-5 = 14
Re (Biology) = 16 – 10 = 6

2. Inclusive Range (Ri) - it is the difference between the upper real limit of the interval
containing the highest observation (Hu) and the lower real limit of the interval
containing the lowest observation (ll).
Ri = Hu - Ll

Example: find the inclusive range for the previous distribution.

Solution:
Ri (Chemistry) = 19.5 - 4.5 = 15
Ri (Biology) = 16.5 - 9.5 = 7

The limitation of the range, as a measure of dispersion, is that it does not provide any
information regarding individual observations. It only gives information on two extreme values.

NOTE: The greater the range, the more heterogeneous is the group and the more spread is the
distribution.

Range is the difference between the highest observation and the

lowest observation.

MEAN OR AVERAGE DEVIATION

The mean or average deviation is used in determining the extent of differences among the
members of a group. It is also an indicator of how compact the group is on a certain measure. It
describes the typical deviation of the scores.

In the succeeding discussions, I will show you how to solve for the mean or average
deviation for a set of observations.

Computations of Mean or Average Deviation

1. Long Method

The mean or average deviation, using the long method, can be calculated using the
following formula:
A.D. = Σ | Χi – X |
N

where: A.D. - Average Deviation

XI - scores
X - mean of the set of scores
N - total number of scores or total frequency
Steps:
1. Find the mean of the set of scores.
2. To get the deviation, subtract the mean (obtained in Step 1) from every score.
3. Add the absolute values of the deviations (obtained in Step 2).
4. Divide the sum (obtained in Step 3) by the total frequency.

Example: Find the average deviation for the test scores presented below.

Student Math Xi - X |Xi - X|

1 14 (14 - 14) = 0 0
2 12 (12 - 14) = -2 2
3 13 (13 - 14) = -1 1
4 16 (16 - 14) = 2 2
5 14 (14 - 14) = 0 0
6 11 (11 - 14) = -3 3
7 14 (14 - 14) = 0 0
8 13 (13 - 14) = -1 1
9 18 (18 - 14) = 4 4
10 15 (15 - 14) = 1 1

Step 1
 Xi = 140
X = 14

Step 2
 (Xi - X) = 0
Step 3

14 =  Xi - X

Step 4
 Xi-X

A.D. = ----------------- = 14 = 1.4

N 10

On the average, the scores of the students in mathematics differ from the mean by 1.4
points. The resulting value of 1.4 serves as an indication of the amount by which the students'
scores in Math are dispersed around the mean of 14. The higher the average deviation, the higher
is the dispersion of the students' scores.

2. For a Score Distribution

The mean of average deviation for a score distribution can be calculated using the
following formula:

|XI - X|f
A.D. = N

Where: A.D. - Average Deviation

XI - scores
X - mean of the set scores
f - frequency of a given score
N - total number of scores or total frequency
Steps:

1. Find the mean of the set of scores.

2. To get the deviation, subtract the mean (obtained in Step 1) from every score. [ Example: 7 *
4 = 28 9 * 10 = 90 ]
3. Multiply the absolute deviation by the corresponding frequency.
[ Example: 7 - 11.3 = - 4.3; 9 - 11.3 = - 2.3]
4. Add the results obtained in Step 3.
5. Divide the sum (obtained in Step 4) by the total frequency.

Example: Find the average deviation for the Math test score distribution.

Step 2
Scores Frequency
(Xi) (f) fXi Xi - X |XI -X| |XI -X|f
14 3 42 0 0 0
12 1 12 -2 2 2
13 2 26 -1 1 2
16 1 16 2 2 2
11 1 11 -3 3 3
18 1 18 4 4 4
15 1 15 1 1 1
N = 10 Fx=140

Step 1 Step 4
 fXI = 140 |XI - X|f = 14

X = 140 = 14
10

Step 4
A.D. = |XI - X|f = 14 = 1.4
N 10

*On the average, the scores vary from the

mean by 1.4 points.

Mean or Average Deviation is the average amount by which the

observations vary from the mean.

STANDARD DEVIATION
The standard deviation is the most important measure of dispersion of a data set. It is
affected by the value of each observation. It is the most stable, and therefore, the most reliable
measure of dispersion.

The standard deviation is the square root of the mean of the squared deviations from the
mean of a distribution.

I will show you how to solve the standard deviation for a set of observations (ungrouped
or grouped).
Computations of Standard Deviation (SD)
1. For Ungrouped Data
The standard deviation for an ungrouped data can be calculated using the following formula:

S.D. = Σ (Χi – X)2

N-1
Where: S.D. - Standard Deviation
XI - scores
X - mean of the set of scores
N - Total number of scores or total frequency

Standard deviation has a minimum value of zero. This means that there is no variation in
the distribution. It increases in values as the variability of the distribution increases.
Standard deviation is useful in comparing distribution of scores. Group of scores with
lower S.D. is more homogeneous than a group of scores with higher S.D.. But S.D. itself has no
obvious or direct significance.

Steps:
1. Find the mean of the set of scores.
2. To get the deviation, subtract the mean (obtained in Step 1) from every score.
3. Get the square of the value obtained in Step 2.
4. Add the results obtained in Step 3.
5. Divide the sum (obtained in Step 4) by N - 1.

Example: Find the standard deviation for the Statistics test scores for the two groups of students

Group I
(Step 2) (Step 3)
Statistics
Scores Xi - X (Xi - X)2
15 (15 - 15) = 0 02 = 0
13 (13 - 15) = -2 (-2)2 = 4
14 (14 - 15) = -1 (-1)2 = 1
17 (17 - 15) = 2 (2)2 = 4
15 (15 - 15) = 0 02 = 0
12 (12 - 15) = -3 (-3)2 = 9
15 (15 - 15) = 0 02 = 0
14 (14 - 15) = -1 (1)2 = 1
19 (19 - 15) = 4 (4)2 = 16
16 (16 - 15) = 1 (1)2 = 1
(Step 1) (Step 4)
XI = 150 (XI - X)2 = 36

X = 150 = 15
10
(Step 5)

S.D. = Σ (Χi – X)2

N-1

= 36
10-1

= 36
9
= 2.00
B. Group II

Students' Scores Xi - X (Xi - X)2

17 (17 - 14.8) = 2.2 4.84
15 (15 - 14.8) = 0.2 0.04
12 (12 - 14.8) = -2.8 7.84
14 (14 - 14.8) = -0.8 0.64
11 (11 - 14.8) = -3.8 14.44
18 (18 - 14.8) = 3.2 10.24
16 (16 - 14.8) = 1.2 1.44
17 (17 - 14. 8) = 2.2 4.84
15 (15 - 14.8) = 0.2 0.04
13 (13 - 14.8) = -1.8 3.24

(Step 1) (Step 4)
XI = 148 (XI - X)2 = 47.60

X = 148 = 14.8
10

(Step 5) Where: N = 10
S.D. = (XI - X)2

N-1

= 47.60
9

= 2.30

Interpretation: The students in Group I are more homogeneous in their scores in Statistics than the
students in Group II. In other words, students in Group I have less varied scores in Statistics
than the students in Group II. This is based on the lower S.D. computed from scores in
Group I (2.00) than in Group II (2.30).

NOTE: The standard deviation of 2.00 or 2.30 when interpreted alone has no physical meaning.
It cannot be considered either as a high or a low value. A value is said to be high or low
relative to another value.

2. For Grouped Data

The standard deviation for a grouped data can be calculated using the following formula:

S.D. =  f(XI - X)2

N-1

Where: S.D. - Standard Deviation

XI - scores
X - mean of the set of scores
f - frequency
N - Total number of scores or total frequency
Steps:

1. Find the mean of the set of scores, applying your knowledge in lesson 1.
2. To get the deviation, subtract the mean (obtained in Step 1) from the midpoint of every
interval.
3. Get the square of the value obtained in Step 2.
4. Multiply the square of the values (obtained in Step 3) with their corresponding frequency.
5. Add the results obtained in Step 4.
6. Divide the sum (obtained in Step 5) by N-1.

Example: Find the standard deviation for the number of passengers in airplanes data are given
below.

Class f
(passengers) Xmid (days) f Xmid Xi - X (Xi - X)2 f (Xi - X)2
50-58 54 2 108 -24.3 590.49 1180.98
59-67 63 5 315 -15.3 234.09 1170.45
68-76 72 14 1008 -6.3 39.69 555.66
77-85 81 18 1458 2.7 7.29 131.22
86-94 90 7 630 11.7 136.89 958.23
95-103 99 4 396 20.7 428.49 1713.96

N=50 Fx= 3915 f (XI - X)2 = 5710.5

(Step 1) X =  fXmid = (108 + 315 + 1008 + 1458 + 630 + 396)

N 50

= 3915 = 78.3
50

S.D. =  f(XI - X)2

N-1

= 5710.5 = 5710.5 = 10.80

50 - 1 49

Standard deviation is the square root of the mean of the squared

deviations of the observations from their mean. It is one of the most
important measures of dispersion.

VARIANCE (s2)

The variance is a statistic closely related to standard deviation. It is the mean of the squared
deviations from the mean. What does that mean? It means that you (1) find the amount by which
each observation deviates from the mean, (2) square those deviations, and (3) find the average of
those squared deviations. You then have "the mean of the squared deviations from the mean." In
other words, the variance (s2) is the square of the standard deviation.
 The variance can be determined using the formula,

s2 = (XI - X)2 (1)

N-1

s2 = N XI2 -  (Xi)2 (2)

N(N - 1)

Where: (s2) = variance

XI - scores
X - mean of the set of scores
N - Total number of scores or total frequency
Example: Find the variance for the previous distribution (passengers in airplanes).

Solution:

S.D. =  (XI - X)2 = 10.8

N-1

s2 = (XI - X)2 = (10.8)2 = 116.64

N-1

Like the standard deviation, the value of variance when interpreted alone has no physical
meaning. When compared with variance of other group with lower value of variance is more
homogeneous. The data gathered from this group are said to be less varied compared to another
group.

Variance is the mean of the squared deviations of the observations

from their mean. It is an important measure for dispersion.