0% found this document useful (0 votes)
20 views11 pages

G4-Measures of Variability

The document discusses various measures of variability that can be used to quantify the dispersion of a data set, including range, variance, standard deviation, interquartile range, quartile deviation, and percentile range. It provides the relevant formulas and examples of calculating each measure for both raw and grouped data.
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
0% found this document useful (0 votes)
20 views11 pages

G4-Measures of Variability

The document discusses various measures of variability that can be used to quantify the dispersion of a data set, including range, variance, standard deviation, interquartile range, quartile deviation, and percentile range. It provides the relevant formulas and examples of calculating each measure for both raw and grouped data.
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/ 11

Engineering Data Analysis Group 4

Measures of Variability

What is Measures of Variability?


– it measures the spread or dispersion of a set of data.

Terms to consider in this topic:

• Range, R
- In a set of data, range refers to the difference between the largest and smallest
measurement.
• Variance of a population
- Denoted by the symbol σ2 .
- The variance of a population of “N” measurements is defined to be the average of
the squares of the deviations of the measurements about the mean μ.

Formula:
∑(𝑥𝑖 − µ)2
σ2 =
𝑁

• Variance of a sample
- Denoted by the symbol 𝑠 2
- The variance of a sample of “n” measurements is defined to be the average of the
squares of the deviations of the measurements about the mean 𝑥̅ divided by (n-1).

Formula:
∑(𝑥𝑖 − 𝑥̅ )2
𝑠2 = 𝑛−1

• Standard Deviation
- The standard deviation of a set of measurements is equal to the positive square root
of the variance.

Formula for the Standard Deviation of a Population:

∑(𝑥𝑖 − µ)2 √𝑁 ∑ 𝑥2 −(∑ 𝑥)2


σ =√2 or σ =
2
𝑁 𝑁
Formula for the Standard Deviation of a Sample:

∑(𝑥− 𝑥̅ )2
𝑠= √
𝑛−1

Example: Ungrouped Data

Consider the sample measurements 7, 2, 5, 3, 4. Find the range and the standard deviation.

Solution:

✓ R=7–2=5
✓ To calculate the standard deviation, follow the steps below.
1. Find the mean.

7+2+5+3+4
𝑥̅ = = 4.2
5
2. Compute the sample variance.
𝑥 (𝑥 − 𝑥̅ ) (𝑥 − 𝑥̅ )2
7 2.8 7.84
2 -2.2 4.84
5 0.8 0.64
3 -1.2 1.44
4 -0.2 0.04
14.8

∑(𝑥𝑖 − 𝑥̅ )2 14.8
𝑠2 = = = 3.7
𝑛−1 4
3. Compute the standard deviation by extracting the square root.
𝑠 = √3.7 = 1.92
Population Variance for Grouped Data
Since there is a considerable number of items in the series, the calculation of the
variance and the standard deviation can be more easily performed if the data are first grouped
into the form of a frequency distribution. Take note of the following processes.
1. The deviation of the midpoint of each group from the arithmetic mean is used as a
measure of the average deviation from the mean of all items in that group.
2. The average deviation of each group is squared to obtain the necessary deviation
squared.
3. The average deviation is multiplied by the frequency indicated for the group in order
to obtain the total of the squared deviations for that group.
4. The totals are added for the entire distribution.
5. Divide the total by N to get the variance. Get the square root of this quotient to get
the standard deviation of the distribution. Below are the formulas for the variances.

∑ 𝑓(𝑥𝑖 −µ)2 ∑ 𝑓𝑥𝑖 2 ∑ 𝑓𝑥𝑖 2


σ2 = or σ2 = − ( )
𝑁 𝑁 𝑁

Example:
Other Method:
However, the computation for the standard deviation may be simplified by the following
method:
a. Instead of using the mean, compute the standard deviation about a conveniently
selected point. For this purpose, the midpoint of any group may be selected.
b. When the class intervals are uniform in size, the calculation may be simplified by
carrying out all computations in the terms of class intervals and then multiplying the
final result by the size of the class interval. The formula for the standard deviation is

𝐶
𝜎= √𝑁 ∑ 𝑓𝑑2 − (𝑓𝑑)2
𝑁

Where σ = standard deviation

d = deviation from the arbitrary mean

f = frequency of values in class interval

C = size of the class interval

Using the same example,

Classes f D fd Fd2
10-19 2 -4 -8 32
20-29 3 -3 -9 27
30-39 5 -2 -10 20
40-49 8 -1 -8 8
50-59 15 0 0 0
60-69 7 1 7 7
70-79 6 2 12 24
80-89 4 3 12 36
-4 154

𝐶
𝜎= √𝑁 ∑ 𝑓𝑑2 − (𝑓𝑑)2
𝑁
10
= √50(154) − (−4)2
50

= (0.2)(87.66)
= 17.53
Sample Variance for Grouped Data
∑ 𝑓 (𝑥𝑖 − 𝑥̅ )2
𝑠2 = 𝑛−1
Other Measures of Dispersions
Interquartile Range

As the dispersion of the frequency distribution is increased, the distance between the quartiles
is enlarge. Since an increased dispersion will be indicated by a greater distance between the quartiles,
the distance may be used as the basis of a measure of dispersion.

𝐼𝑄𝑅 = 𝑄3 − 𝑄1

The Quartile Deviation or Semi-Interquartile Range

One half of the distance between the quartiles may be used as a measure of the average
distance of each quartile from the median. This value is used as a measure of dispersion.
𝑄3 − 𝑄1
𝑄𝐷 =
2
The 10-90 Percentile Range

In a similar manner as the dispersion is increased, the distance between the various percentiles
will be increased. The distance between the two percentiles may thus be used as a measure of
dispersion.

10 – 90 Percentile Range = = 𝑃90 − 𝑃10


CamScanner
CamScanner
CamScanner
CamScanner
CamScanner

You might also like