Professional Documents
Culture Documents
Range
Variance
Standard Deviation
Coefficient of Variation
Summary Definitions
Example:
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14
Range = 13 - 1 = 12
Measures of Dispersion:
Why The Range Can Be Misleading
7 8 9 10 11 12 7 8 9 10 11 12
Range = 12 - 7 = 5 Range = 12 - 7 = 5
Percentile Range
90
P10 P90 P10
Quartile Deviation
Q Q3 Q1
The inter-quartile range is frequently reduced to the
measure of semi-interquartile range, known as the
quartile deviation (QD), by dividing it by 2. Thus
Q3 Q1
QD
2
Mean Deviation
f i xi x
MDx i 1
n
k = Number of classes
xi= Mid point of the i-th class
fi= frequency of the i-th class
Standard Deviation
(X i X) 2
S i1
n -1
For A Population:
The Standard Deviation σ
i
(X μ) 2
σ i1
N
Approximating the Standard Deviation
from a Frequency Distribution
Assume that all values within each class interval are
located at the midpoint of the class
s
(x x) 2
f
n -1
Where n = number of values or sample size
x = midpoint of the jth class
f = number of values in the jth class
Measures of Dispersion:
The Standard Deviation
Sample
Data (Xi) : 10 12 14 15 17 18 18 24
n=8 Mean = X = 16
x ẍ (x - ẍ) (x - ẍ)2
∑x = ∑(x - ẍ)2 =
ẍ = ∑x/n =
∑(x - ẍ)2/n =
√∑(x - ẍ)2/n =
x ẍ (x - ẍ) (x - ẍ)2
5
9
3
2
7
9
8
2
2
3
∑x =
∑(x - ẍ)2 =
ẍ = ∑x/n =
∑(x - ẍ)2/n =
√∑(x - ẍ)2/n =
x ẍ (x - ẍ) (x - ẍ)2
5
9
3
2
7
9
8
2
2
3
∑x = 50
∑(x - ẍ)2 =
ẍ = ∑x/n = 50/10 = 5
∑(x - ẍ)2/n =
√∑(x - ẍ)2/n =
x ẍ (x - ẍ) (x - ẍ)2
5 5
9 5
3 5
2 5
7 5
9 5
8 5
2 5
2 5
3 5
∑x = 50
∑(x - ẍ)2 =
ẍ = ∑x/n = 50/10 = 5
∑(x - ẍ)2/n =
√∑(x - ẍ)2/n =
x ẍ (x - ẍ) (x - ẍ)2
5 5 0
9 5 4
3 5 -2
2 5 -3
7 5 2
9 5 4
8 5 3
2 5 -3
2 5 -3
3 5 -2
∑x = 50
∑(x - ẍ)2 =
ẍ = ∑x/n = 50/10 = 5
∑(x - ẍ)2/n =
√∑(x - ẍ)2/n =
x ẍ (x - ẍ) (x - ẍ)2
5 5 0 0
9 5 4 16
3 5 -2 4
2 5 -3 9
7 5 2 4
9 5 4 16
8 5 3 9
2 5 -3 9
2 5 -3 9
3 5 -2 4
∑x = 50
∑(x - ẍ)2 =
ẍ = ∑x/n = 50/10 = 5
∑(x - ẍ)2/n =
√∑(x - ẍ)2/n =
x ẍ (x - ẍ) (x - ẍ)2
5 5 0 0
9 5 4 16
3 5 -2 4
2 5 -3 9
7 5 2 4
9 5 4 16
8 5 3 9
2 5 -3 9
2 5 -3 9
3 5 -2 4
∑x = 50
∑(x - ẍ)2 = 80
ẍ = ∑x/n = 50/10 = 5
∑(x - ẍ)2/n =
√∑(x - ẍ)2/n =
x ẍ (x - ẍ) (x - ẍ)2
5 5 0 0
9 5 4 16
3 5 -2 4
2 5 -3 9
7 5 2 4
9 5 4 16
8 5 3 9
2 5 -3 9
2 5 -3 9
3 5 -2 4
∑x = 50
∑(x - ẍ)2 = 80
ẍ = ∑x/n = 50/10 = 5
∑(x - ẍ)2/n = 8
√∑(x - ẍ)2/n =
x ẍ (x - ẍ) (x - ẍ)2
5 5 0 0
9 5 4 16
3 5 -2 4
2 5 -3 9
7 5 2 4
9 5 4 16
8 5 3 9
2 5 -3 9
2 5 -3 9
3 5 -2 4
∑x = 50
∑(x - ẍ)2 = 80
ẍ = ∑x/n = 50/10 = 5
∑(x - ẍ)2/n = 8
√∑(x - ẍ)2/n =
2.8°C
Why?
Standard deviation
tells us the average
distance of each
score from the mean.
68% of normally
distributed data is
within 1 sd each side
of the mean
95% within 2 sd
Almost all is within 3
sd
Example
Mean IQ = 100, sd = 15
What is the IQ of 68% of
population (ie what is the
range of possible IQs)?
Between what IQ scores
would 95% of people be?
Dan says he has done an
online IQ test, and he
has an IQ of 170. Should
you believe him?
Why/not?
Another example
S i1
n -1
Where X = arithmetic mean
n = sample size
Xi = ith value of the variable X
Another formula for Variance
Sample Variance s 2
x 2
f
x 2
X = arithmetic mean
n = sample size
Xi = ith value of the variable X
f = frequency
For A Population:
The Variance σ2
Average of squared deviations of values from
the mean
N
Population variance: (X i μ) 2
σ2 i1
N
S
CV
X
The Coefficient of Variation
CV
This can be used to compare two distributions directly
to see which has more dispersion because it does not
depend on units of the distribution.
Measures of Dispersion:
Comparing Coefficients of Variation
Stock A:
Average price last year = $50
Standard deviation = $5
S $5
CVA 100%
100% 10%
X $50 Both stocks
Stock B: have the same
standard
Average price last year = $100
deviation, but
Standard deviation = $5 stock B is less
variable relative
to its price
S $5
CVB 100%
100% 5%
X $100
Chap 3-40
Sample statistics versus
population parameters
The less the data are spread out, the smaller the
range, variance, and standard deviation.
If the values are all the same (no variation), all these
measures will be zero.