At the end of the lesson, students should be able to:

• identify the different measures of variation

• characterize and formulate each measure of
variation
• relate variance to standard deviation
• solve problems on measures of variation
The data below shows the results (in months) of two brands of outdoor paint how long
they last before fading.

Brand A Brand B
10 35
60 45
Brand A 50 30 Brand B
30 35
40 40
20 25

The mean for Brand A is 35 months The mean for Brand B is 35 months
Do both brands of paint last equally well?

10 20 30 35 40 50 60

brand B Brand A
performs more the spread, and
consistently. variation, is quite
different

10 20 30 35 40 50 60
Brand B
MEASURES OF VARIABILITY
❖Range Range = highest value – lowest value

𝑅𝑎𝑛𝑔𝑒 𝑜𝑓 𝐵𝑟𝑎𝑛𝑑 𝐴 = 60 − 10 = 50 𝑚𝑜𝑛𝑡ℎ𝑠

10 35 60

𝑅𝑎𝑛𝑔𝑒 𝑜𝑓 𝐵𝑟𝑎𝑛𝑑 𝐵 = 45 − 25 = 𝟐𝟎 𝒎𝒐𝒏𝒕𝒉𝒔

25 45
35
MEASURES OF VARIABILITY
❖Variance the average of the squares of the distance each value is
from the mean

population variance, 𝜎 2
σ 𝑥−𝜇 2
2
𝜎 = where x = individual values
𝑁 µ = population mean
N = population size
sample variance, 𝑠 2
σ 𝑥 − 𝑥ҧ 2
2 where x = individual values
𝑠 = 𝑥ҧ = sample mean
𝑛−1
n = sample size
MEASURES OF VARIABILITY
❖ Standard Deviation square root of the variance
population standard deviation, σ

σ 𝑥−𝜇 2
𝜎= 𝜎2 =
𝑁

sample standard deviation, s

σ(𝑥 − 𝑥)ҧ 2
𝑠= 𝑠2 =
𝑛−1
Variance and Standard Deviation for the data set of Brand A paint
Brand A: 10, 60, 50, 30, 40, 20
➢ Solve first for the mean.
10 + 60 + 50 + 30 + 40 + 20
𝜇=
6
210
=
6
𝜇 = 35
➢ Subtract the mean for each data value.
x x-µ
10 10 – 35 = -25
60 60 – 35 = 25
50 50 – 35 = 15
30 30 – 35 = -5
40 40 – 35 = 5
20 20 – 35 = -15
Variance and Standard Deviation for the data set of Brand A paint
➢ Square each result.

x x-µ (𝒙 − 𝝁)𝟐
10 10 – 35 = -25 (−25)2 = 625
60 60 – 35 = 25 252 = 625
50 50 – 35 = 15 152 = 225
30 30 – 35 = -5 (−5)2 = 25
40 40 – 35 = 5 52 = 25
20 20 – 35 = -15 (−15)2 = 225
1750

➢ Find the sum of the squares.

625 + 625 + 225 + 25 + 25 + 225 = 1750
Variance and Standard Deviation for the data set of Brand A paint
➢ Divide the sum by N to get the variance.
1750
𝑣𝑎𝑟𝑖𝑎𝑛𝑐𝑒: 𝜎2 =
6
𝜎 2 = 291.67

➢ Take the square root of the variance to get the standard deviation.

𝑠𝑡𝑎𝑛𝑑𝑎𝑟𝑑 𝑑𝑒𝑣𝑖𝑎𝑡𝑖𝑜𝑛: 𝜎 = 291.67

σ = 17.08
The Unbiased Estimator
The unbiased estimator of the population variance is an estimator whose value approximates the
expected value of population variance. It is denoted by 𝑠 2 , and its formula is

σ 𝑥 − 𝑥ҧ 2
2 where 𝑥ҧ = 𝑠𝑎𝑚𝑝𝑙𝑒 𝑚𝑒𝑎𝑛
𝑠 =
𝑛−1 𝑛 = 𝑠𝑎𝑚𝑝𝑙𝑒 𝑠𝑖𝑧𝑒

The standard deviation of a sample (denoted by s) is σ 𝑥 − 𝑥ҧ 2

𝑠= 𝑠2 =
𝑛−1
Shortcut Formula for s2 and s

2
ቀσ𝑥) 2
2
σ𝑥 − 2
൫σ𝑥 )
𝑛 σ𝑥 −
𝑠2 = 𝑠= 𝑛
𝑛−1 𝑛−1
For a vacant position, the manager interviewed the following numbers of applicants
each day over a five-day period: 16, 19, 15, 15, and 14. Find the variance and
standard deviation.
Let us use the shortcut formula for s2 and s
x 𝒙𝟐
2
16 256 ቀσ𝑥)
2
σ𝑥 −
19 361 𝑠2 = 𝑛
𝑛−1
15 225
79 2
15 225
1263 −
14 196 𝑠2 = 5 = 3.7
5−1
σx = 79 σ𝑥 2 = 1263
𝑠 = 3.7 = 1.92
Variance and Standard Deviation for Grouped Data σ 𝑓𝑥i 2
σ 𝑓𝑥i2
−
σ𝑓
𝑠2=
σ𝑓 − 1
Classes frequency 𝒙𝒊 𝒇𝒙𝒊 𝒇𝒙𝒊 𝟐 𝑓𝑥𝑖 2 = 𝑓𝑥𝑖 ∙ 𝑥𝑖
6 - 10 1 8 8 64
13 52
𝑓𝑥01 2 = 8 ∙ 8 = 64
11 - 15 4 676
16 - 20 6 18 108 1944 𝑓𝑥02 2 = 52 ∙ 13 = 676
21 - 25 10 23 230 5290 𝑓𝑥03 2 = 108 ∙ 18 = 1944
26 - 30 8 28 224 6272 𝑓𝑥03 2 = 230 ∙ 23 = 5290
31 - 35 5 33 165 5445
𝑓𝑥04 2 = 224 ∙ 28 = 6272
36 - 40 2 38 76 2888
σf = 36 σfxi= 863 22579 𝑓𝑥05 2 = 165 ∙ 33 = 5445
𝑓𝑥06 2 = 76 ∙ 38 = 2888
 σ 𝑓𝑥i 2
2
σ 𝑓𝑥i −
2 σ𝑓
𝑠 =
σ𝑓 − 1
2
863
22579 − 36
𝑠2 =
36 − 1

22579 − 20688.03
𝑠2 =
35
𝑠 2 = 54.03

𝑠 = 54.03 = 7.35
Uses of Variance and Standard Deviation
➢ to determine the spread of the data
If the variance or standard deviation is large, the data is more dispersed. This information is
useful in comparing two or more data sets to determine which is more variable.

➢ to determine the consistency of the variable

If the variance or standard deviation is small, the variable is most likely dependable.

➢ to determine the number of data values that fall within a specified interval in a
distribution.
For any distribution, at least 75% of the data values will fall within two standard deviations of
the mean.

➢ used quite often in inferential statistics.