Measures of Dispersion

AGBS | Bangalore
 Summary Measures
Describing Data Numerically

Central Tendency Quartiles Variation Shape

Arithmetic Mean Range Skewness

Median Interquartile Range

Mode Variance

Geometric Mean Standard Deviation

Coefficient of Variation
 Quartiles
 Quartiles split the ranked data into 4 segments with
an equal number of values per segment

25% 25% 25% 25%

Q1 Q2 Q3

 The first quartile, Q1, is the value for which 25% of the
observations are smaller and 75% are larger
 Q2 is the same as the median (50% are smaller, 50% are
larger)
 Only 25% of the observations are greater than the third
quartile
 Quartile Formulas

Find a quartile by determining the value in the

appropriate position in the ranked data, where

First quartile position: Q1 = n/4

Second quartile position: Q2 = n/2 (the median position)

Third quartile position: Q3 = 3n/4

where n is the number of observed values

 Quartiles

 Example: Find the first quartile

Sample Data in Ordered Array: 11 12 13 16 16 17 18 21 22

(n = 9)
Q1 is in the 9/4 = 2.25 position of the ranked data

so Q1 = 12.25

Q1 and Q3 are measures of noncentral location

Q2 = median, a measure of central tendency
 Quartiles
(continued)
 Example:
Overtime 10-15 15-20 20-25 25-30 30-35 35-40 Total
Hours
No. of 11 20 35 20 8 6 100
employees

Calculate Median, First Quartile, 7th Decile & Interquartile Range

 Measures of Variation
Variation

Range Interquartile Variance Standard Coefficient

Range Deviation of Variation

 Measures of variation give

information on the spread
or variability of the data
values.

Same center,
different variation
 Range

 Simplest measure of variation

 Difference between the largest and the smallest
values in a set of data:

Range = Xlargest – Xsmallest

Example:

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14

Range = 14 - 1 = 13
 Disadvantages of the Range
 Ignores the way in which data are distributed

7 8 9 10 11 12 7 8 9 10 11 12
Range = 12 - 7 = 5 Range = 12 - 7 = 5

 Sensitive to outliers
1,1,1,1,1,1,1,1,1,1,1,2,2,2,2,2,2,2,2,3,3,3,3,4,5
Range = 5 - 1 = 4

1,1,1,1,1,1,1,1,1,1,1,2,2,2,2,2,2,2,2,3,3,3,3,4,120
Range = 120 - 1 = 119
 Coefficient of Range

Coefficient of Range = [L-S]/[L+S]

Where,
L = Largest value in the data set
S = Smallest value in the data set
 Interquartile Range

 Can eliminate some outlier problems by using

the interquartile range

 Eliminate some high- and low-valued

observations and calculate the range from the
remaining values

 Interquartile range = 3rd quartile – 1st quartile

= Q3 – Q1
 Interquartile Range

Example:
Median X
X Q1 Q3 maximum
minimum (Q2)
25% 25% 25% 25%

12 30 45 57 70

Interquartile range
= 57 – 30 = 27
 Coefficient of QD

Coefficient of QD= [Q3-Q1]/[Q3+Q1]

 Average Deviation

Coefficient of AD = (Average Deviation)/(Mean or Median)

 Calculate the Coefficient of AD
(mean)
Sales (in thousand Rs) No. of days

10-20 3

20-30 6

30-40 11

40-50 3

50-60 2
 Variance

 Average (approximately) of squared deviations

of values from the mean
n
 Sample variance:
 (X i  X) 2

S 2 i 1
n -1
Where X = mean
n = sample size
Xi = ith value of the variable X
 Standard Deviation
 Most commonly used measure of variation
 Shows variation about the mean
 Is the square root of the variance
 Has the same units as the original data
 Sample standard deviation:
n

 i
(X  X ) 2

S i 1
n -1
 Calculation Example:
Sample Standard Deviation
Sample
Data (Xi) : 10 12 14 15 17 18 18 24
n=8 Mean = X = 16

(10  X ) 2  (12  X ) 2  (14  X ) 2    (24  X ) 2

S
n -1

(10  16) 2  (12  16) 2  (14  16) 2    (24  16) 2


7

130 A measure of the “average”

  4.308 scatter around the mean
7
Population vs. Sample Standard
Deviation
Population vs. Sample Standard
Deviation (Grouped Data)
Calculate the Standard Deviation
for the following sample
Number of pins knocked down in ten-pin bowling matches
 Solution
Number of pins knocked down in ten-pin bowling matches
 Calculate the Standard Deviation
for the following sample
Heights of students
Solution
 Calculate the Standard Deviation
for the following sample
Number of typing errors
 Comparing Standard Deviations

Data A
Mean = 15.5
11 12 13 14 15 16 17 18 19 20 21 S = 3.338
Data B
Mean = 15.5
11 12 13 14 15 16 17 18 19 20 21 S = 0.926
Data C
Mean = 15.5
11 12 13 14 15 16 17 18 19 20 21 S = 4.567
 Measuring variation

Small standard deviation

Large standard deviation

Advantages of Variance and
Standard Deviation

 Each value in the data set is used in the

calculation

 Values far from the mean are given extra

weight
(because deviations from the mean are squared)
 Coefficient of Variation

 Measures relative variation

 Always in percentage (%)
 Shows variation relative to mean
 Can be used to compare two or more sets of
data measured in different units

 S 
C V     100%

 X 
 Comparing Coefficient
of Variation
 Stock A:
 Average price last year = $50

 Standard deviation = $5

S $5
 
C V A     100%   100%  10%
X $50 Both stocks
have the same
 Stock B:
standard
 Average price last year = $100 deviation, but
stock B is less
 Standard deviation = $5 variable relative
to its price
S $5
C VB     100%   100%  5%
X $100
Sample vs. Population CV
 Calculate!

 Calculate the coefficient of variation for

the following data set.
 The price, in cents, of a stock over five trading days
was 52, 58, 55, 57, 59.
Solution
Pooled Standard Deviation
Pooled Standard Deviation
 Z Scores

 A measure of distance from the mean (for example, a

Z-score of 2.0 means that a value is 2.0 standard
deviations from the mean)
 The difference between a value and the mean, divided
by the standard deviation
 A Z score above 3.0 or below -3.0 is considered an
outlier

XX
Z
S
 Z Scores
(continued)

Example:
 If the mean is 14.0 and the standard deviation is 3.0,
what is the Z score for the value 18.5?

X  X 1 8 .5  1 4 .0
Z    1 .5
S 3 .0
 The value 18.5 is 1.5 standard deviations above the
mean
 (A negative Z-score would mean that a value is less
than the mean)
 Shape of a Distribution

 Describes how data are distributed

 Measures of shape
 Symmetric or skewed

Left-Skewed Symmetric Right-Skewed

Mean < Median Mean = Median Median < Mean
 The Empirical Rule

 If the data distribution is approximately

bell-shaped, then the interval:
 μ  1σ contains about 68% of the values in
the population or the sample

68%

μ
μ  1σ
 The Empirical Rule
 μ  2σ contains about 95% of the values in
the population or the sample
 μ  3σ contains about 99.7% of the values
in the population or the sample

95% 99.7%

μ  2σ μ  3σ
 Chebyshev Rule

 Regardless of how the data are distributed,

at least (1 - 1/k2) x 100% of the values will
fall within k standard deviations of the mean
(for k > 1)
 Examples:
At least within
(1 - 1/12) x 100% = 0% ……..... k=1 (μ ± 1σ)
(1 - 1/22) x 100% = 75% …........ k=2 (μ ± 2σ)
(1 - 1/32) x 100% = 89% ………. k=3 (μ ± 3σ)
 Calculate!
Variable Frequency
44-46 3
46-48 24
48-50 27
50-52 21
52-54 5
 Calculate the mean & the standard dev.
 Calculate the percentage of
observations between the mean &  2σ
 Find the correct Mean & Std. dev.
The Mean & Standard Deviation of a set of 100
observations were worked out as 40 & 5
respectively. It was later learnt that a mistake had
been made in data entry – in place of 40 (the
correct value), 50 was entered.
 Skewness
 A fundamental task in many statistical analyses is to
characterize the location and variability of a data set
(Measures of central tendency vs. measures of
dispersion)
 Both measures tell us nothing about the shape of the
distribution
 It is possible to have frequency distributions which differ
widely in their nature and composition and yet may
have same central tendency and dispersion.
 Therefore, a further characterization of the data
includes skewness
 Positive & Negative Skew

 Positive skewness
 There are more observations below the mean than
above it
 When the mean is greater than the median
 Negative skewness
 There are a small number of low observations and a
large number of high ones
 When the median is greater than the mean
 Measures of Skew

 Skew is a measure of symmetry in the

distribution of scores
Normal
(skew = 0)

Positive Negative Skew

Skew
Measures of Skew
 The Rules

I. Rule One. If the mean is less than the median,

the data are skewed to the left.

II. Rule Two. If the mean is greater than the

median, the data are skewed to the right.
Karl Pearson Coefficient of Skewness
 Karl Pearson Coefficient - Properties

Advantage – Uses the data completely

Disadvantage – Is sensitive to extreme values
 Calculate!

 Calculate the coefficient of skewness & comment on

its value
 Calculate!
Profits (lakhs) No. of Companies
100-120 17
120-140 53
140-160 199
160-180 194
180-200 327
200-220 208
220-240 2

 Calculate the coefficient of skewness & comment on

its value
Bowley’s Coefficient of Skewness
 Bowley’s Coefficient - Properties

Advantage – Is not sensitive to extreme values

Disadvantage – Does not use the data completely
Calculate!
Kelly’s Coefficient of Skewness
 Calculate!
Wages No. of Workers
Below 200 10
200-250 25
250-300 145
300-350 220
350-400 70
Above 400 30

 Obtain the limits of daily wages of the central 50% of the

workers
 Calculate Bowley’s & Kelly’s Coefficients of Skewness
Moments about the Mean