UNIT 6 MEASURES OF SKEWNESS AND

KURTOSIS
Structure
Objectives
Introduction
Concept of Skewness
6.2.1 Karl Pearson's Measure of Skewness
6.2.2 Bowley's Measure of Skewness
6.2.3 Kelly's Measure of Skewness
Moments
Concept and Measure of Kurtosis
Let Us Sum Up
Key Words
Some Useful Books
6.8 Answers or Hints to Check Your Progress Exercises

6.0 OBJECTIVES
After going through this Unit, you will be able ta :
distinguish between a symmetrical and a skewed distribution;
compute various coefficients to measure the extent of skewness in a
distribution;
distinguish between platykurhc, mesokurtic and leptokurtic distributions; and
compute the coefficient of kurtosis.

6.1 INTRODUCTION
In this Unit you will learn various techniques to distingush between various shapes
of a frequency distribution. This is the final Unit with regard to the summarisation
of univariate data. This Unit will make you familiar with the concept of skewness
and kurtosis. The need to study these concepts arises fiom the fact that the
measures of central tendency and dispersion fail to describe a distribution
completely. It is possible to have fkquency distributionswhich differ widely in their
nature and composition and yet may have same central tendency and dispersion.
Thus, there is need to supplement the measures of central tendency and dispersion.
Consequently, in t h ~ Unit,
s we shall discuss two such measures, viz, measures of
skewness and kurtosis.

6.2 CONCEPT OF SKEWNESS

The skewness of a distribution is defined as the lack of symmetry. In a symmetrical
distribution, the Mean, Median and Mode are equal to each other and the
ordinate at mean divides the distribution into two equal parts such that one
S u ~ n ~ n a r i s n t i o un f part is mirror image of the other (Fig. 6.1). If some observations, of very
Cnivariate Data
high (low) magnitude, are added to such a distribution, its right (left) tail gets
elongated.

I Symmetrical Distribution

Fig. 6.1

, Positively Skewed Distribution I Negatively Skewed Distribution

Fig. 6.2

These observations are also known as extreme observations. The presence of

extreme observations on the right hand side of a distribution makes it positively
skewed and the three averages, viz., mean, median and mode, will no longer be
equal. We shall in fact have Mean > Median > Mode when a distribution is
positively skewed. On the other hand, the presence of extreme observations to
the left hand side of a distribution make it negatively skewed and the relationship
between mean, median and mode is: Mean < Median < Mode. In Fig. 6.2 we
depict the shapes of positively skewed and negatively skewed distributions.

The direction and extent of skewness can be measured in various ways. We shall
discuss four measures.@skewness in this Unit.
6.2.1 Karl ~ e a r h n ' sMeasure of Skewness
In Fig. 6.2 you noticed that the mean, median and mode are not equal in a skewed
distribution. The Karl Pearson's measure of skewness is based upon the
divergence of mean from m o b in a skewed distribution.

Since Mean = Mode in a symmetrical distribution, (Mean - Mode) can be taken

as an absolute measure of skewness. The absolute measure of skewness for a
distribution depends upon the unit of measurement. For example, if the mean =
2.45 qetre and mode = 2.14 metre, then absolute measure of skewness will be
2.45 aetre - 2.14 metre = 0.31 metre. For the same distribution, if we change
the d t of measurement to centimetres, the absolute measure of skewness is 245
centimetre - 2 14 centimetre = 3 1 centimetre. In order to avoid such a problem Measures Skewness and
Kurtosis
Karl Pearson takes a relative measure of skewness.

A relative measure, independent of the units of measurement, is defined as the

Karl Pearson b Coeficient of Skewness Sk,given by
Mean - Mode
S, =
s.d.

The sign of Skgives the direction and its magnitude gives the extent of skewness.

If Sk> 0, the distribution is positively skewed, and if S, < 0 it is negatively skewed.

So far we have seen that Skis strategicallydependent upon mode. If mode is not
defined for a distribution we cannot find Sk.But empirical relation between mean,
median and mode states that, for a moderately symmetrical distribution, we have
Mean - Mode = 3 (Mean - Median)

Hence Karl Pearson's coefficient of skewness is defined in t m s of median as

Example 6.1: Compute the Karl Pearson's coefficient of skewness fiom the
following data:
Table 6.1

Height (in inches) Number of Persons

58 10
59 18
60 30
61 42
62 35
63 28
64 16
65 8

Table for the computation of mean and s.d.

Height (X) u =X - 61 No. of persons V ) fi fu2

58 -3 10 - 30 90
59 -2 18 -36 72
60 -1 30 - 30 30
61 0 42 0 0
62 1 35 35 35
63 2 28 56 112
64 3 16 48 144
65 4 8 32 128

Total 187 75 611

Summarisation of
Univariate Data
Mean =
75
61 + -= 61.4
187

To find mode, we note that height is a continuous variable. It is assumed that the
height has been measured under the approximation that a measurement on height
that is, e.g., greater than 58 but less than 58.5 is taken as 58 inches while a
measurement greater than or equal to 58.5 but less than 59 is taken as 59 inches.
Thus the given data can be written as

Height (in inches) No. of persons

57.5 - 58.5
58.5 - 59.5
59.5 - 60.5
60.5 - 61.5
61.5 - 62.5
62.5 - 63.5
63.5 - 64.5
64.5 - 65.5
By inspection, the modal class is 60.5 - 61.5. Thus, we have

12
. Mode = 60.5 + 12+7 x 1 = 61.13

61.4 - 61.13 -
- 0.153.
Hence, the Karl Pearson's coeficient of skewness Sk =

Thus the distribution is positively skewed.

6.2.2 Bowley's Measure of Skewness
This measure is based on quartiles. For a symmetrical distribution, it is seen that
Q, and Q3are equidistant ftom median. Thus (Q3- Md)- (M,- Q,)can be taken
as an absolute measure of skewness.

A relative measure of skewness, known as Bowley's coefficient (SQ),is given

by
The Bowley's coefficient for the data on heights given in Table 6.1 is computed Measures of Skewness
Kurtosis
below.

Height (in inches) No. of persons V) Cumulative Frequency

57.5 - 58.5 10 10-
58.5 - 59.5 18 28
59.5 - 60.5 30 58
60.5 - 61.5 42 100
61.5 - 62.5 35 135
62.5 - 63.5 28 163
63.5 - 64.5 16 179
64.5 - 65.5 8 187

Computation of Q, :

N
Since p = 46.75, the first quartile class is 59.5 - 60.5. Thus

la, = 59.5, C = 28, fa,= 30 and h = 1.

Computation of M, (Q,) :

N
Since 2 = 93.5, the median class is 60.5 - 61.5. Thus

Im = 60.5, C = 58, fm = 42 and h = 1.

Computation of Q, :
3N
Since = 140.25, the third quartile class is 62.5 - 63.5. Thus
la, = 62.5, C = 135, f Q = 28 and h = 1.

62.688 - 2 x 61.345 + 60.125 = 0.048 .

Hence, Bowley's coefficient SQ =
62688 - 60.125

6.2.3 Kelly's Measure of Skewness

Bowley's measure of skewness is based on the middle 50% of the observations
because it leaves 25% of the observaticins on each extreme of the distribution.
As an improvement over Bowley's measure, Kelly has suggested a measure
based on P,,and,P,,so that only 10% of the observations on each extreme are
ignored.
Summarisation of Kelly's coefficient of skewness, denoted by S, is given by
Univariate Data

Note that P,,= M, (median).

The value of S,, for the data given in Table 6.1, can be computed as given
below.

Computation of P,,:

Since - - lo ls7
loN - = 18.7. 10th percentile lies in the class 58.5 - 59.5. Thus
100 100

Computation of P, :

Since W N --
100 100
= 168.3, 90th percentile lies in the class 63.5 - 64.5. Thus

lpw = 63.5,C = 163,fpm = 16 and h = 1.

P, = 63.5+ - 163 x 1= 63.831.

16

Hence, Kelly's coefficient

It may be noted here that although the coefficient S,, So and S, are not
comparable, however, in the absence of skewness, each of them will be equal to
zero.

Check Your Progress 1

1) Compute the Karl Pearson's coefficient of skewness from the following

data :
Daily Expenditure (Rs.) : 0-20 20-40 40-60 60-80 80-100
No. of families : 13 25 27 19 16

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
 Measures of Skewness and
2) The following figures relate to the size of capital of 285 companies : Kurtosis
Capital (in Ks. lacs.) 1-5 610 11-15 1620 21-25 2630 31-35 ibtal
No. of companies 20 27 29 38 48 53 285

Compute the Bowley's and Kelly's coefficients of skewness and interpret

the results.

3) The following measures were computed for a frequency distribution :

Mean = 50, coefficient of Variation = 35% and
Karl Pearson's Coefficient of Skewness = - 0.25.
Compute Standard Deviation, Mode and Median of the distribution.

6.3 MOMENTS
The rth moment about mean of a distribution, denoted by p,, is given by

P h N i=1 e - where r = 0, 1, 2, 3, 4, .........

Thus, rth moment about mean is the mean of the rth power of deviations of
observations from their arithmetic mean. In particular,
1 "
if r = 0, we have PO =-~h(xi
N i=1 -x)O= I ,

if r = 1, we have PI xi -x)=o,
1 "
N ,=1

1 "
l - - ~ f ; ( x-zy
if r = 2, we h a v e ~ =
N i=l . i = a 2 ,

if r = 3, we have P 3 = L?h
N i=1 (xi - x)' and so on.
Summarisation of In addition to the above, we can define raw moments as moments about any
Univariate Data
arbitrary mean.

Let A denote an arbitrary mean, then uth moment about A is defined as

When A = 0, we get various moments about origin.

Moment Measure of Skewness
The moment measure of skewness is based on the property that, for a symmetrical
distribution, all odd ordered central moments are equal to zero.

We note that p, = 0, for every distribution, therefore, the lowest order moment
that can provide an absolute measure of skewness 'is p3.

Further, a coefficient of skewness, independent of the units of measurement, is

given by

CI 3=
a3=-
0
+& p, and y, are defined as the first beta and first
= Y 1 , where

gamma coefficients respectively. P, is measure of kurtosis as you will come to

know in the next Section.
CL; 1
Very often, the skewness is measured in terms of P 1 = 3 ,
where the sign of
F12
iI
skewness is determined by the sign of p,.

Example 6.2: Compute the Moment coefficient of skewness (P,) from the
following data.
Marks Obtained : 0-10 10-20 20-30 30-40 40-50 50-60 60-70 !
Frequency : 6 12 22 24 16 12 8

Table for the computations of mean, s.d. and p,. I

X-3 5
Class Frequency Mid- u=- fu fu2 fu3
10
Intervals (f ) values ( X )
0 - 10 6 5 -3 - 18 54 - 162
10 - 20 12 15 -2 - 24 48 - 96
20 - 30 22 25 -1 - 22 22 -22
30 - 40 24 35 0 0 0 0
40 - 50 16 45 1 16 16 16
50 - 60 12 55 2 24 48 96
60 - 70 8 65 3 24 72 216

Total 100 0 260 48

Since Xfu = 0, the mean of the distribution is 35.

The second moment p, is equal to the variance (oZ) and its positive square root
is equal to standard deviation (a).
3lcasures of Skewness a n d
Kurtosis I
260
p 2 =-~100=260, and
100

Since the sign of p3is positive and p, is small, the distribution is slightlypositively
skewed.

If the mean of a distribution is not a convenient figure like 35, as in the above
example, the computation of various central moments may become a cumbersome
task. Alternatively, we can first compute raw moments and then convert them into
central moments by using the equations obtained below.

Conversion of Raw Moments. into Central Moments

We can write

1 "
= -Eh[(xi
N
-A)-piIr 1
(Since p i = G Z h ( X i - ~ ) = z - ~ )

Expanding the term within brackets by binomial theorem, we get

From the above, we can write

pr = - f C I p ~ - l p+i r ~ 2 p : - 2 p i 2 -r C 3 p ; - 3 p 1, 3 +""'.

In particular, taking r = 2, 3, 4, etc., we get

p 2 = p ; - 2 ~ l p ; 2 + 2 ~ 2 p ; p=~p2i - p i 2 (since p ; = 1)
Summarisation of Example 6.3: Compute the first four moments about mean from the following
Univariate Data
data.
ClassIntervals : 0 - 1 0 10-20 20-30 30-40
Frequency V) : 1 3 4 2

Table for computations of raw moments (Take A = 25).

X-25
Class Intervals f Mid-Value u= -
10
fu fu2 fu3 fu4
0
0 - 10 1 5 -2 -2 4 - 8 16
10 - 20 3 15 - 1 - 3 3 -3 3
20 - 30 4 25 0 0 0 0 0
30 - 40 2 35 1 2 2 2 2

Total 10 -3- 9 -9 21

From the above table, we can write

-9x lo3
pi= I(, = 9 0 0 and

Moments about Mean

By definition,

Check Your Progress 2

1) Calculate the first four moments about mean for the following distribution.
Also calculate 9, and comment upon the nature of skewness.
Marks : 0 - 20 20 - 40 40 - 60 60 - 80 80 - 100 '
Frequency : 8 28 35 17 12
 Measures of Skewness and
Kurtosis

2) The first three moment of a distribution about the value 3 of a variable are
2,10 and 30respectively. Obtain F, p2,p3and hence P,. Comment upon
the nature of skewness.

6.4 CONCEPT AND MEASURE OF KURTOSIS

Kurtosis is another measure of the shape of a distribution. Whereas skewness
measures the lack of symmetry of the frequency curve of a distribution, kurtosis
is a measure of the relative peakedness of its fi-equency curve. Various frequency
curves can be divided into three categories depending upon the shape of their peak.
The three shapes are termed as Leptokurtic, Mesokurtic and Platykurhc as shown
in Fig. 6.3.

Leptokurtic
Mesokurtic
Platykurtic

Fig. 6.3
Summarisation o f
Univarinte Data P4
A measure of kurtosis is given by P 2 = 2,
a coefficient given by Karl Pearson.
P2
The value of p2 = 3 for a mesokurtic curve. When P2 > 3, the curvt: is more
peaked than the mesokurtic curve and is tenned as leptokurtic. Similarly, when
p2 < 3 , the curve is less peaked than the mesokurtic curve and is called as
platykurtic curve.

Example 6.4: The first four central moments of a distribution are 0,2.5,0.7 and
18.75. Examine the skewness and kurtosis of the distribution.

To examine skewness, we compute p,.

Since p3> 0 and p, is small, the distribution is moderately positively skewed.

18.75 - 3.0 ,
= --
P4
Kurtosis is given by the coefficient P2 = ---i-
P2 (q2
Hence the curve is mesokurtic.

Check Your Progress 3

1) Compute the first four central moments h m the following data. Also find the
two beta coefficients.
Value 5 10 15 20 25 30 35
Frequency : 8 15 20 32 23 17 5

2) The first four moments of a distribution are 1,4, 10 and 46 respectively.

Compute the moment coefficients of skewness and kurtosis and comment upon
the nature of the distribution.
 Measures of Skewness and
6.5 LET US SUM UP Kurtosis

In this Unit you have learned about the measures of skewness and kurtosis. These
two concepts are used to get an idea about the shape of the fiequency curve of
a distribution. Skewness is a measure of the lack of symmetry whereas kurtosis is a
measure of the relative peakedness of the top of a fiequency curve.

6.6 KEY WORDS

Skewness: Departure from symmetry is skewness.

Moment of Order r: It is defined as the arithmetic mean of the rth power of

deviations of observations.

Coefficient of Kurtosis: It is a measure of the relative peakedness of the top

of a frequency curve.

6.7 SOME USEPUL BOOKS

Elhance, D. N. and V. lhance, 1988, Fundamentals of Statistics, Kitab Mahal,
Allahabad.
Nagar, A. L. and R. K. Dass, 1983, Basic Statistics, Oxford University Press,
DeIhi .

Mansfield, E., 1991, Statistics for Business and Economics: Methods and
Applications, W.W. Norton and Co.
Yule, G U. and M. G Kendall, 1991, An Introduction to the Theov of Statistics,
Universal Books, Delhi.

6.8 ANSWERS OR HINTS TO CHECK YOUR

PROGRESS EXERCISES
Check Your Progress 1
1) 0.237
2) - 0.12, - 0.243
3) 17.5, 54.38, 51.46
Check Your Progress 2

1) 0,499.64, 2579.57, 589111.61, 0.053, skewness is positive.

2) 5, 6, -14,0.907, since p3is negative the distribution is negatively skewed.
Check Your Progress 3

1) 0,59.99, - 50.18, 8356.64,0.012 (negatively skewed), 2.32 (platykurtic).

2) 0,3. Thus thedistribution is symmetrical and mesokurtic. Such a distribution
is also known as a Normal Distribution.