Scribd
Upload
11 views

MEASURES OF SHAPE

The document discusses measures of shape in statistics, specifically skewness and kurtosis. It explains how to calculate the Pearson Coefficient of skewness and provides examples demonstrating positive and negative skewness, as well as how to compute kurtosis using the fourth moment about the mean. Additionally, it includes activities for calculating skewness and kurtosis based on sample data.

Uploaded by

Jayce
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
11 views

MEASURES OF SHAPE

The document discusses measures of shape in statistics, specifically skewness and kurtosis. It explains how to calculate the Pearson Coefficient of skewness and provides examples demonstrating positive and negative skewness, as well as how to compute kurtosis using the fourth moment about the mean. Additionally, it includes activities for calculating skewness and kurtosis based on sample data.

Uploaded by

Jayce
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 18

MEASURES OF SHAPE

• Measure of Skewness
• Measure of Kurtosis
The direction of the skewness of a distribution is based on the relationship
between mean and the median. If the mean is higher than the median, the
curve is skewed to the right (positively skewed) but skewed to the left
(negatively skewed) if it is the reverse. The measure of skewness is called the
Pearson Coefficient of skewness (SK), expressed in the following formula:

𝟑(𝑴𝒆𝒂𝒏−𝑴𝒆𝒅𝒊𝒂𝒏) 𝒙− 𝒙෥ )
𝟑(ഥ
𝑺𝑲 = 𝒐𝒓 𝑺𝑲 =
𝑺𝒕𝒂𝒏𝒅𝒂𝒓𝒅 𝑫𝒆𝒗𝒊𝒂𝒕𝒊𝒐𝒏 𝑺

If the coefficient of skewedness is negative, the distribution is skewed to the


left, while if positive, it is skewed to the right. The amount of magnitude of the
coefficient of skewness will indicate the extent of the skewness of the curve.
In symbols, if the SK= 0, the distribution is symmetric; if SK>0, skewed to the
right (positively skewed); and if SK<0, skewed to the left (skewed negatively).
From example 1, we have the variance 𝑆 2 = 13.71, therefore the
standard deviation 𝑆 = 13.71 = 3.7
108
The mean 𝑥ҧ = = 13.5 𝑎𝑛𝑑 𝑡ℎ𝑒 𝑀𝑒𝑑𝑖𝑎𝑛 𝑥෤ = 13
8

3(𝑀𝑒𝑎𝑛 −𝑀𝑒𝑑𝑖𝑎𝑛)
The coefficient og skewness: 𝑆𝐾 =
𝑆𝑡𝑎𝑛𝑑𝑎𝑟𝑑 𝐷𝑒𝑣𝑖𝑎𝑡𝑖𝑜𝑛

3(13.5−13)
𝑆𝐾 = = 0.41
3.7
Since 𝑆𝐾 > 0, the distribution is skewed to the right or positively
skewed.
For Grouped Data
Example 2: The variance is 𝑆 2 = 52.93, then standard deviation is
𝑆 = 52.93 = 7.26

4,131
The mean, 𝑥ҧ = = 47.48 and the median is
87

43.5−33
𝑥෤ = 45.5 + 5 = 47.6 , therefore the coefficient of
25
skewness
3(47.38−47.6)
𝑆𝐾 = = −0.09
7.26
Since 𝑆𝐾 < 0, the distribution is skewed to the left of negatively
skewed.
Example 1: determine the measure of kurtosis using the fourth
moment about the mean for the following 7 raw sample data: 3, 7,
10, 14, 18, 21, and 25.
σ 𝑥−𝑥ҧ 2 372
Using the formula for variance, we have 𝑆 2 = = = 62
𝑛−1 7−1
Since 𝑆 2 = 62, 𝑡ℎ𝑒 𝑠 4 = 62 2 = 3844

Thus, we shall obtain the measure of kurtosis using the fourth


moment about the mean:

σ 𝑥−𝑥ҧ 4 34,596
𝐾𝑢 = = = 1.29
𝑛𝑠4 7(3,844)
Hence, the curve is classified as platykurtic since 𝐾𝑢 < 3.
For Grouped Data
Example 2: calculate the measure of coefficient of kurtosis using the
fourth moment from the mean of the frequency distribution of the
production sales of a sample of 80 salesmen of a manufacturing
company.
Solving the mean and variance:

σ 𝑓𝑥 3,840 σ 𝑓 𝑥 − 𝑥ҧ 2
2
5,000
𝑥ҧ = = = 48 𝑎𝑛𝑑 𝑆 = = = 63.29
𝑛 80 𝑛−1 79

If 𝑆 2 =63.29. then 𝑠 4 = 63.29 2 = 4,005.62

Thus, the measure of kurtosis using fourth moment from the mean
is:
σ 𝑓 𝑥−𝑥ҧ 4 725,000
𝐾𝑢 = = = 2.26
𝑛𝑠 4 80(4,005.62)
Since the measure of kurtosis is less than 3, then the curve of the
frequency distribution is classified as platykurtic.
Activity 1
1. Given the following 12 production outputs per day of employees
in a slipper manufacturing company, compute the Personian
coefficient of skewness and the measure of kurtosis on the fourth
moment about the mean. Assume these data as sample data.
35, 28, 40, 35, 43, 24, 38, 20, 29, 45, 37, 30
2. A survey of the monthly salaries of 88 executives in 10 insurance
companies revealed the following data.

You might also like