Skewness

Download as pdf or txt
Download as pdf or txt
You are on page 1of 11

Understanding Skewness

Data Science and A.I. Lecture Series

Bindeshwar Singh Kushwaha

PostNetwork Academy
Reach PostNetwork Academy

Website
PostNetwork Academy | www.postnetwork.co

YouTube Channel
www.youtube.com/@postnetworkacademy

PostNetwork Academy Facebook Page


www.facebook.com/postnetworkacademy

LinkedIn Page
www.linkedin.com/company/postnetworkacademy

Bindeshwar Singh Kushwaha (PostNetwork Academy) Understanding Skewness 2/7


Introduction
What is Skewness?
When mean and variance are the same in two different datasets, the distributions of the data could
still be very different in terms of their shape, particularly with respect to their symmetry and the
spread of data points around the central value. This is where skewness becomes crucial for further
understanding the underlying distributions.

Bindeshwar Singh Kushwaha (PostNetwork Academy) Understanding Skewness 3/7


Introduction
What is Skewness?
When mean and variance are the same in two different datasets, the distributions of the data could
still be very different in terms of their shape, particularly with respect to their symmetry and the
spread of data points around the central value. This is where skewness becomes crucial for further
understanding the underlying distributions.

Frequency Distributions having Same Mean and Standard Deviation


Consider the following distributions:
Variable (X) 0-5 5-10 10-15 15-20 20-25 25-30
Frequency (Y) 10 30 60 30 10 0

Variable (X) 0-5 5-10 10-15 15-20 20-25 25-30


Frequency (Y) 10 5 10 20 30 10
In both distributions, the mean is 15 and the standard deviation is 6.
Bindeshwar Singh Kushwaha (PostNetwork Academy) Understanding Skewness 3/7
Frequency Polygons

Bindeshwar Singh Kushwaha (PostNetwork Academy) Understanding Skewness 4/7


Frequency Polygons


Their shapes differ, as seen in the frequency polygons:
Variable (X) 0-5 5-10 10-15 15-20 20-25 25-30 Variable (X) 0-5 5-10 10-15 15-20 20-25 25-30
Frequency (Y) 0 5 10 20 30 10 Frequency (Y) 0 5 10 20 30 10

80 100 90
70 90
80
Frequency

Frequency
60 60
60 70
50 60
40 30 30
50
30 40 30 30
30 20
20 10 10 20 10 10
10 10
5 10
15 20 25 30 5 10 15 20 25 30
Variable Variable
The first distribution is symmetrical, while the second distribution is asymmetrical or skewed.

Bindeshwar Singh Kushwaha (PostNetwork Academy) Understanding Skewness 4/7


Example Distributions

Definitions of Skewness
Definition 1: “Skewness refers to the asymmetry or lack of symmetry in the shape of a frequency
distribution.” — Morris Hamberg

Bindeshwar Singh Kushwaha (PostNetwork Academy) Understanding Skewness 5/7


Example Distributions

Definitions of Skewness
Definition 1: “Skewness refers to the asymmetry or lack of symmetry in the shape of a frequency
distribution.” — Morris Hamberg
Definition 2: “A distribution is said to be ‘skewed’ when the mean and the median fall at different
points in the distribution, and the balance (or centre of gravity) is shifted to one side or the other -
to left or right.” — Garret

Bindeshwar Singh Kushwaha (PostNetwork Academy) Understanding Skewness 5/7


Example Distributions

Definitions of Skewness
Definition 1: “Skewness refers to the asymmetry or lack of symmetry in the shape of a frequency
distribution.” — Morris Hamberg
Definition 2: “A distribution is said to be ‘skewed’ when the mean and the median fall at different
points in the distribution, and the balance (or centre of gravity) is shifted to one side or the other -
to left or right.” — Garret
Definition 3: “When a series is not symmetrical it is said to be a asymmetrical or skewed.” —
Croxton & Cowden

Bindeshwar Singh Kushwaha (PostNetwork Academy) Understanding Skewness 5/7


Types of Skewness

Positive or Right Skew


Customer Spending (Positive Skew)
100
In a positively skewed 100
distribution, the right tail is Frequency
90
80
longer than the left tail. The 80
70
majority of data values are 70
concentrated on the left side 60
of the distribution, while a 50
50

few larger values stretch the 40


distribution to the right. 30
30
Example : Most customers at 20
20
a retail store make small 10
10 Spending5 ($)
purchases, but a few make 0

very large purchases. 50 100 150 200 300 500

Bindeshwar Singh Kushwaha (PostNetwork Academy) Understanding Skewness 6/7


Types of Skewness
Negative of Left Skew
Exam Scores (Negative Skew)
100
In a negatively skewed Frequency 90
distribution, the left tail is longer 90 85
or fatter than the right tail. The 80

majority of data values are


80
70
concentrated on the right side of 70
the distribution, while a few
smaller values stretch the 60
50
distribution to the left. 50
Example : In a class, most
students score very well on a 40
relatively easy exam, but there are 30

a few students who perform 30


poorly. The distribution of scores 20
will likely be negatively skewed 12
8
because the low scores drag the 10 4
6

average down.
2 Score 0(%)
10 20 30 40 50 60 70 80 90 100
Bindeshwar Singh Kushwaha (PostNetwork Academy) Understanding Skewness 7/7

You might also like