Professional Documents
Culture Documents
Variability
1
Consider the two sets of data below.
Set A
25, 28, 28, 30, 30, 33, 35, 40, 41, 45
Set B
10, 15, 23, 28, 28, 30, 39, 45, 52, 65
2
Range
Example:
In Set A, the range is 45 – 25 = 20.
In Set B, the range is 65 – 10 = 55.
3
4
Mean Absolute Deviation (MAD)
Ungrouped Data
xi x
MAD
N
Where
xi = score
x = mean of the scores
N = total number of scores
5
Mean Absolute Deviation (MAD)
Grouped Data
MAD
f X X
N
Where
X = class mark
X = mean
f = frequency
N = total number of cases
6
Population Variance
Ungrouped Data
2 i 1
N
7
Sample Variance
Given a random sample x1, x2, …, xn, the
sample variance is
n 2
xi x
2
Biased estimator: s i1
n
n 2
xi x
2
Unbiased estimator: s i1
n 1
8
Sample Standard Deviation
n 2
xi x
s i 1
n 1
9
Computational Formula for the Sample Variance
(unbiased)
2 2
2
n x x
s
nn 1
10
Example:
Set A
25, 28, 28, 30, 30, 33, 35, 40, 41, 45
We have x 33.5 , n 10
10
xi 33.5
2
s
2 i 1
10 1 9
s 2 43
s 7
11
Example:
Set B
10, 15, 23, 28, 28, 30, 39, 45, 52, 65
We have x 33.5 , n 10
10
xi 33.5
2
s 287 and s 17
2
12
Sample Variance from Grouped Data
n 2
f xi x
2
s i1
n
where
f = frequency
x = class mark
x = mean
n = total number of observations
13
Computational Formula for the Sample Variance
from Grouped Data
2 2
2
n f x f x
s 2
n
where
f = frequency
x = class mark
n = total number of observations
14
Measure of Relative Variation
To compare the variability of data sets
measured in different units, we use the
measure of relative variation called coefficient
of variation. This index expresses the standard
deviation as a percentage relative to the mean.
It’s value is given by
s
C .V . 100%
x
15
Example:
Determine which data set is more spread
out.
16
We first compute the means and standard
deviations of the sets of data.
Data Set 1:
x 24 years and s 3.742 years
Data Set 2 :
x P 8875 and s P 2267.984
17
So, we have
Data Set 1:
s 3.742 years
C.V . 100% 100% 15.59%
x 24 years
Data Set 2 :
s P 2267.984
C.V . 100% 100% 25.55%
x P 8875
Therefore, net take home pay is more scattered wrt the
mean than years of teaching experience of teachers.
18
Measure of Skewness and Kurtosis
19
Computing for Skewness
20
Kurtosis is a measure of the
“tailedness” of the probability distribution.
A standard normal distribution has
kurtosis of 3 and is recognized as mesokurtic.
An increased kurtosis (>3) can be visualized as
a thin “bell” with a high peak whereas a
decreased kurtosis corresponds to a
broadening of the peak and “thickening” of
the tails. Kurtosis >3 is recognized as
leptokurtic and <3 as platykurtic (lepto=thin;
platy=broad).
21
Computing for Kurtosis
22
Illustration on Kurtosis
23