UNIT 11 COMPUTATION OF MEASURES

OF VARIABILITY*
Structure
11.0 Objectives
11.1 Introduction
11.2 Computing Different Measures of Variability
11.2.1 Range (R)
11.2.2 Quartile Deviation (QD)
11.2.2.1 Calculation of Quartile Deviation for Ungrouped Data
11.2.2.2 Calculation of Quartile Deviation for Grouped Data
11.2.3 Average Deviation (AD) or Mean Deviation (MD)
11.2.3.1 Computation of Average Deviation for Ungrouped Data
11.2.3.2 Calculation of Average Deviation for Grouped Data
11.2.4 Standard Deviation (SD)
11.2.4.1 Calculation of Standard Deviation for Ungrouped Data
11.2.4.2 Computations of SD from Grouped Data by Long Method
11.2.4.3 Calculation of SD from Grouped Data by Short Method

11.3 Let Us Sum Up

11.4 References
11.5 Answers to Check Your Progress
11.6 Unit End Questions

11.0 OBJECTIVES
After reading this unit, you will be able to:
 compute range;
 compute quartile deviation, for ungrouped and grouped data;
 compute average deviation, for ungrouped and grouped data; and
 compute standard deviation, for ungrouped and grouped data and with the
help of short method.

11.1 INTRODUCTION
In the previous two units, we discussed about measures of central tendency and
measures of variability. We discussed that average like mean, median and
mode condense the series into a single figure. These measures of central
tendency tell us something about the general level of magnitude of the
distribution but they fail to show anything further about the distribution. It is
not fully representative of a population unless we know the manner in which
the individual items scatter around it. A further description of the series is
necessary if we are to gauge how representative the average is. To cite an

* Dr. Usha Kulshreshtha, Faculty, Psychology, University of Rajasthan, Jaipur. 197

Statistical example, in a country the average income may be very high, yet there may be
Methods-I great disparity in its distribution among people. As a result, a majority of the
people may be living below the poverty line. When we want to make
comparison between two groups, it is seen that at times the value of the means
is the same in both the groups but there is a large difference between individual
participants in the groups. This difference amongst the participants within the
same group is termed as variation, that is within the groups the participants vary
a great deal even though they have the same means. Therefore to make
accurate and meaningful comparisons between groups, we should use
variability along with central tendency. In the last unit we have learned about
the concept of variability, different measures of variability, their merits,
limitations and uses. In this unit, we will learn how to compute the range,
quartile deviation, average deviation and standard deviation.

11.2 COMPUTING DIFFERENT MEASURES OF

DISPERSION (VARIABILITY)
As explained in the last unit that there are four measures of computing
variability or dispersion within a set of scores:
1) Range(R)
2) Quartile Deviation (QD)
3) Average Deviation (AD) or Mean Deviation (MD)
4) Standard Deviation (SD)
Each of the above measures of variability give us the degree of variability or
dispersion by the use of a single number and tells us how the individual scores
are spread throughout the distribution. In the following sections, we will
discuss the methods of computation of the above measures of dispersion.

11.2.1 Range(R)
Range is the difference between the highest and the lowest score for a group of
participants whose scores are given.
The formula for Range is as follows:
R=H-L
Where,
H=Highest scores in the distribution
L=Lowest score in the distribution
Let us understand the steps in computation of range with the help of an
example,
For example, if there are 10 students who have obtained marks in history as
mentioned below:
50, 45, 42, 46, 55, 54, 59, 60, 62, 64

198
Step 1: Arrange the scores in ascending order. Computation of
Measures of
42, 45, 46, 50, 54, 55, 59, 60, 62, 64 Variability

Step 2: Identify the lowest and the highest score in the data
In the above data, the lowest score is 42 and the highest score is 64.
Step 3: Compute range with the help of the following formula:
R=H-L
64–42=22.
Thus, the range obtained is 22.

11.2.2 Quartile Deviation (QD)

The inter quartile range is a measure of dispersion and is equal to the
difference between the third and first quartiles. Half of the inter quartile range
is called semi interquartile range or Quartile Deviation. The formula for
computation of QD is
QD = Q3 – Q1/2

Where,
Q1 = first quartile of the data

Q3= third quartile of the data

Quartile is an additional way of disaggregating data. Each quartile represents

one-fourth of an entire population or the group. The quartile deviation has an
attractive feature that the range "median+QD" contains approximately 50% of
the data. The quartile deviation is also an absolute measure of dispersion. Its
relative measure is called coefficient of quartiled deviation or semi-inter
quartile range. It is defined by the relation;
Coefficient of quartile deviation = (Q3-Q1) / (Q3+Q1)

The quartile deviation is one half the scale distance between the 75th and 25th
percentile in a frequency distribution. The 25th percentile or Q1 is the first
quartile on the score scale, the point below which lies 25% of the scores.The
75th percentile or Q3 is the third quartile on the score scale, the point below
which lie 75% of the scores. To find quartile deviation, we must first compute
Q3 and Q1.

There are grouped data and ungrouped data in all cases and thus to compute
quartile deviation, we have to find out first if it is a grouped data ore
ungrouped data. We will first see how the quartile deviation is calculated
from ungrouped data.
11.2.2.1 Calculation of Quartile Deviation for Ungrouped Data
Let us understand steps in computation of quartile deviation for ungrouped
data with the help of the following example:

199
Statistical The scores obtained by students in Psychology class test are :
Methods-I
24, 25, 23, 26, 29, 30, 27, 35, 34, 36, 28
Step 1: Arrange the data in ascending order.
23, 24, 25, 26, 27, 28, 29, 30, 34, 35, 36
Step 2: Compute Q1
Q1 = (N+1)/4th position
N = 11
Q1 = 11+1/4th position = 3rd position = 25
Step 3: Compute Q3
Q3 = 3(N+1)/4th position
N = 11
Q3 = 3(11+1)/4 = 9th position = 34
Step 4: Compute QD with help of the following formula
QD = Q3 – Q1/2

In the case of our data Q3 is 34 and Q1/2 is 25

QD =34–25/2
= 9/ 2
= 4.5
11.2.2.2 Calculation of Quartile Deviation for Grouped Data
From the grouped data, Quartile Deviation can be computed with the following
formula:
QD = Q3 – Q1/2

Further,
Q1 = l+i[(N/4-cum fi)]/fq
Q3 = l+i[(3N/4-cum fi)]/fq
Where,
l = the exact lower limit of the interval in which the quartile falls.
i = the length of the interval
cum fi = cumulative f up to interval which contains the quartile
fq = the f on the interval containing the quartile.
Let us understand steps in computation of quartile deviation for grouped data
with the help of the following example:

200
 Computation of
Class intervals Frequencies Cumulative Measures of
(f) frequencies Variability
195-199 1 50
190-194 2 49
185-189 4 47
180-184 5 43
175-179 8 38
170-174 10 30
165-169 6 20
160-164 4 14
155-159 4 10(1+3+2+4)
150-154 2 6 (1+3+2)
145-149 3 4 (1+3)
140-144 1 1

Step 1: Q1 is calculated using the formula given below:

Q1 = l+i[(N/4–cumfi)]/fq
To locate the Q1, we take N/4. In the above example N/4=12.5

l = 159.5 (50/4 = 12.5 th item from down below) (falls in 160-164)

fi = 10 cumulated scores upto interval containing Q1

fq = 4, the f on the interval on which Q1 falls

i = 5 (Class Interval)
Substituting in formula we have,
Q1=159.5+5{(12.5-10)}/4=162.62
Step 2: To calculate the third quartile, that is Q3.
Q3 = l+i [(3N/4–cum fi)]/fq
To locate the Q3 we take 3xN/4 of our scores. In the above example,
3N/4 is 3×50/4 = 37.5
3/4N = 37.5 (37.5th item that falls in 175-179)
l = 174.5 is the exact lower limit of interval which contains Q3
Cum fi = 30, sum of scores upto interval which contains Q3

i=5
fq=8
Q3=174.5+5(37.5-30)/8 =179.19

Step 3: Finally, substituting in formula, we have the quartile deviation as given

below in the formula; 201
Statistical QD = Q3–Q2 / 2
Methods-I
Q = {(179.19)–-(162.62)}/2
= 8.28
Thus, quartile deviation for the above data is 8.28.

11.2.3 Average Deviation (AD)

Average deviation as well can be computed for ungrouped and grouped data.
11.2.3.1 Computation of Average Deviation for Ungrouped Data
In case of ungrouped data, the average deviation is calculated by the
following formula,
AD=∑│x│/N
Where,
∑│x│= Total of deviation from mean
N= Total number of observations
Let us understand the steps in computation of average deviation for ungrouped
data with the help of example given below:
Following are the scores obtained by five students on a test:
Students Scores Deviation from the
Mean | x |
1 6 -4
2 8 -2
3 10 0
4 12 2
5 14 4
Total=50 Total=12(Ignore signs)
Mean=50/5=10

Step 1: Compute mean with the help of formula M=∑X/N

M=50/5 =10
Step 2: Compute deviation from the mean as has been computed in the third
column in above example.
It is seen from the table, that the deviations (x) are =0,-2,-4, +2, +4
Step 3: Compute total for these deviations without considering the + and –
signs. the total is obtained as 12.
Step 4: Use the formula to compute average deviation.
AD = ∑│ x│/N
= 12/5
=2.4.
202
Thus, the average deviation is obtained as 2.4. Computation of
Measures of
11.2.3.2 Calculation of Average Deviation for Grouped Data Variability

The average deviation for grouped data can be computed by the following
formula,
AD= ∑│fx│/N
Where,
∑│fx│ =Add all the fx without considering the + and – sign
N= Number of observations
The above formula and calculation of AD can be illustrated by the example
given below.
Let us understand the steps in computation of average deviation for grouped
data with the help of example given below:

Class Frequency Mid fX x=M-X fx

Interval (f) Point
(X)
(1) (2) (3) (4) (5) (6)
110-114 3 112 336 20.33 60.99
(112×3) (112- (20.33×3)
91.67=20.33)
105-109 4 107 428 15.33 61.32
100-104 6 102 612 10.33 61.98
95-99 8 97 776 5.33 42.64
90-94 15 92 1380 .33 4.95
85-89 10 87 870 –4.67 –46.67
80-84 7 82 574 –9.67 –67.69
75-79 4 77 308 –14.67 –58.68
70-74 3 72 216 –19.67 –59.01
Total= 60 Total= Total =
5500 463.93
Mean=9
1.67

Step 1: Identify midpoints of the class interval and mention them in column
three, as can be seen above.
Step 2: Multiply respective frequencies and mid-points as shown in column 4.
Step 3: For obtained fX, compute mean with the help of formula M=∑fX/N.
M=5500/60
=91.67
Step 4: x is then computed with the formula x=M–X.
203
Statistical Step 5: Average deviation is then computed with the help of following
Methods-I formula:
AD=∑│fx│/N
=463.93/60
=7.73
Thus, the average deviation obtained is 7.73.

11.2.4 The Standard Deviation (SO)

Standard Deviation is the most stable measure of variability. Therefore, it is
most commonly used in research studies. Standard deviation can be computed
for ungrouped and grouped data.
11.2.4.1 Calculation of Standard Deviation (SD) for Ungrouped Data
Standard deviation for ungrouped data can be computed by the following
formula:
SD =√∑x2/N
The above formula can be explained by the following example.
Let us understand the steps in computation of standard deviation for ungrouped
data with the help of example given below:

Scores Deviation from the Deviation square

(X) mean (x) (x²)
52 -8 64
50 -10 100
56 -4 16
68 8 64
65 5 25
62 2 4
57 -3 9
70 10 100
Total = 480 ∑x2= 382
Mean= 60
Step 1: Add all the scores (∑X) and divide this sum by the number of scores
(N) and find out mean. Mean of the given scores is
M = ∑X/N
=480/8
=60.
Step 2: Find out deviation x by computing X – x, as given in second column
above.
Step 3: Square all the deviation to get x2.
Step 4: Add all the squared deviation to get ∑x2.
204
Step 5: Compute standard deviation with the help of the formula: Computation of
Measures of
SD= √(∑x2) /N Variability

= √ 382/8
=√ 47.7
=6.91
Thus, the standard deviation for this data is 6.91.
11.2.4.2 Computations of SD for Grouped Data by Long Method
Standard deviation of grouped data can be computed by the formula,
SD= √ ∑fx2/N
Where,
∑fx2= when frequencies (f) are multiplied with their respective devation
squared (x2), fx2 is obtained total of all fx2 is fx2
N= Total number of scores
Let us understand the steps in computation of standard deviation for grouped
data with the help of example given below:
Class Frequency Midpoint fX Deviation of Deviation fx²
interval (f) (X) midpoint from squared
the mean (x²)
(x= X- M)
(1) (2) (3) (4) (5) (6) (7)
127-129 1 128 128 17.6 309.76 309.76
124-126 2 125 250 14.6 213.16 426.32
121-124 2 122 244 11.6 134.56 269.12
118-120 2 119 238 8.6 73.96 147.92
115-117 4 116 464 5.6 31.36 125.44
112-114 4 113 452 2.6 6.76 27.04
109-111 4 110 440 –0.4 0.16 0.64
106-108 2 107 214 –3.4 11.56 23.12
103-106 2 104 208 –6.4 40.96 81.92
100-102 2 101 202 –9.4 88.36 176.72
Total= 25 ∑fx= 2760 ∑fx²= 1588

Step 1: Midpoint is computed for respective class intervals and entered in

column 3 as shown in the above table.
Step 2: fX is then computed by multiplying the frequencies and the midpoint,
the values thus obtained are entered in column 4.
Step 3: Add all the scores under fX and divide this sum by the number of
scores (N) and find out mean. Thus, M= ∑fX/ N, that is, 2760/ 25= 110.4. The
mean obtained is 110.4.
Step 4: x is now computed by subtracting the M from X (midpoint). The
values thus obtained are entered in column 5.
205
Statistical Step 5:The x is then squared to obtain x² (column 6).
Methods-I
Step 6: fx² is then computed by multiplying f and x². The values thus obtained
are entered in column 7.
Step 7: Add the fx2 to obtain ∑fx². In the present example it is obtained as
1588.
Step 8: Compute standard deviation with help of the formula:
SD= √ ∑fx2/N
= √1588/ 25
= √63.52
= 7.97
The standard deviation thus obtained is 7.97.
11.2.4.3 Calculation of SD for Grouped Data by Short Method
Standard deviation from grouped data can also be computed by the following
formula.
SD= i √ ∑fx′2/N-(∑(fx′)/N)²
where,
i= The size of the class interval
∑fx/2= when frequencies are multiplied with their respective deviations of
midpoint from the mean fx′ is obtained square of fx′ is fx′2 and total of all the
fx′2 is fx′2
Let us understand the steps in computation of standard deviation for grouped
data by short method with the help of example given below:
Class Frequency Midpoint Deviation of (fx′) fx′²
interval (f) (X) midpoint from the
mean
(x′= X- AM/i)
(1) (2) (3) (5) (6) (7)
127-129 1 128 4 4 16
124-126 2 125 3 6 36
121-124 2 122 2 4 16
118-120 2 119 1 2 4
115-117 4 116 0 0 0
112-114 4 113 -1 -4 16
109-111 4 110 -2 -8 64
106-108 2 107 -3 -6 36
103-106 2 104 -4 -8 64
100-102 2 101 -5 -10 100
Total= 25 ∑fx′ = - ∑fx′²=
20 352

Step 1: Find out midpoint of each class interval, that is entered in column 3.
Step 2: Assume one value as mean. In this example, assumed mean is taken as
206 116.
Step 3: Find out the difference between midpoint and assumed mean and Computation of
Measures of
divide it by class intervals to get x′, (x′= X- AM/i) and enter the obtained value in
Variability
column 5.
Step 4: Multiply each x′ by respective frequency and get fx′ (column 6).
Step 5: Multiply fx′ is squared to get fx′2. Add all the fx′2 to obtain ∑fx′.
Step 6: Compute standard deviation with the help of the formula
SD= i √ ∑fx′2/N-(∑(fx′)/N)²
= 3 √352/ 25- (-20/ 25)²
= 3 √14.08 - (-0.8)²
= 3 √ 14.08 - 0.64
= 3 √3.44
= 3 x 1.85
= 5.55
The standard deviation thus obtained is 5.55.
Check Your Progress 1
1) What is the formula for range?

2) List the steps in computation of standard deviation for ungrouped data.

......................................................................................................................
......................................................................................................................
......................................................................................................................
......................................................................................................................
......................................................................................................................

11.3 LET US SUM UP

In this unit, we covered the computation of different measures of variability,
like the range, average deviation, quartile deviation and standard deviation.
Computation of each of the measures of dispersion was discussed with the
help of steps and examples.

11.4 REFERENCES
Garrett, H.E. (1981), Statistics in Psychology and Education, (Tenth edition),
Bombay, Vakils Feffer and Simons Ltd.
McBride, Dawn M. (2018). The Process of Statistical Analysis in Psychology.
207
Statistical Sage. USA
Methods-I
Minium, E.W., King, B.M. & Bear. G (2001). Statistical Reasoning in
Psychology and Education (3rd edition), Singapore, John Wiley &Sons, Inc.
Mohanty, B. & Misra, Santa (2016). Statistics for Behavioural and Social
Sciences. Sage. New Delhi.

11.5 ANSWERS TO CHECK YOUR PROGRESS

Check Your Progress 1
1) What is the formula for range?
R=H-L
2) List the steps in computation of standard deviation for ungrouped data.
The steps in computation of standard deviation for ungrouped data are
Step 1: Add all the scores (∑x) and divide this sum by the number of scores
(N) and find out mean.
Step 2: Find out deviation x by computing X- x.
Step 3: Square all the deviation to get x2.
Step 4: Add all the squared deviation to get ∑x2.
Step 5: Compute standard deviation with the help of the formula.

11.6 UNIT END QUESTIONS

1) Compute the range, average deviation and standard deviation from the
following ungroupeddata:
a) 30, 35, 36, 39, 42, 46, 38, 34, 35
b) 52, 50, 56, 68, 65, 62, 57, 70
2) Calculate quartile deviationforthefollowingscores:
6, 3, 9, 9, 5, 7, 9, 6, 8, 4, 8, 5, 7, 9, 3, 2, 9, 5, 7
3) Calculate Average deviation of the following scores:
Class Interval Frequency
40-44 3
35-39 4
30-34 6
25-29 12
20-24 7
15-19 5
10-14 1

208
4) Calculate the Quartile deviation and Standard Deviation for the following Computation of
Measures of
frequency distribution:
Variability
Scores Frequency
70-71 2
68-69 2
66-67 3
64-65 4
62-63 6
60-61 7
58-59 5
56-57 1
54-55 2
52-53 3
50-51 1

209
Statistical
Methods-I

210 blank

210