Descriptive Statistics - Part 3

* Measures of dispersion express quantitatively the degree of
variation or dispersion of values in a population or in

a sample. It measures how spread out a set of data is.
* Along with measures of central tendency, measures of
dispersion are widely used in practice as descriptive
statistics.
* Some measures of dispersion are the variance , standard
deviation, the range , the coefficient of variation and the
interquartile range.
Measures of Dispersion I (looks at the
spread of the data)
3
Description Applicability Advantage Disadvantage
Range Difference between the largest 1. Interval/ratio 1. Simple to calculate 1. Highly influenced
and the smallest value in the 2. No outliers exist by outliers.
data. 2. Does not use all
data
Variance/ Variance is the average 1. Interval/ratio 1. Use all the data 1. Not resistant to
standard squared deviations from the 2. When no outliers outliers.
deviation mean. exist 2. Variance depends
on the units of
Standard deviation is square measurement,
root of the variance. therefore not easy
Commonly used. to make
comparisons.
Coefficient Measures variability relative 1. Interval/ratio 1. Ensures fairness in 1. Not

resistant to
of Variation to the magnitude of the data. 2. When no outliers comparison outliers.
It compares the variability of exist 2. Use all the data
different data
3
Measures of Dispersion II
4
Raw data Ungroup frequency Group frequency Notations

distribution distribution
Range Max. value – Min Same NA NA

value
X = the actual values (for raw
Variance/  x  2
 fx 2
 fx 2 data and ungrouped freq.
standard x 2

n
 fx 2

n
 fx 2

n
dist.)
deviation s2  s2  s2  = midpoints (for group
n 1 n 1 n 1 freq. dist.)
f = frequency
n = sample size
 = summation or sum of
s 2 = sample variance
Coefficient of
variation  stdev  Same as raw data Same as raw data
CV    100%
 mean 
4
A computer salesperson, X, sells the following number of
computers in 12 months:
34, 47, 1, 15, 57, 24, 20, 11, 19, 50, 28, 37.
calculate the:
a.Mean
b.Variance
c.Standard deviation
d.Coefficient of variation
The data below shows the age distribution of a sample of persons
from Country X in 1990.
Age (years) No. of Persons

Less than 20 10
20 - <30 14
30 - <40 25
40 – <50 27
Calculate the mean,
50 – <60 15
standard deviation and
60 and over 9
the coefficient of
variation for the data and
interpret your findings.
Age group Number unemployed
15-19 3 688
20-24 4 031
25-34 5 432

35-44 4 360
45-54 3 162
55-64 1 702
Calculate the standard

deviation and coefficient
of variation.
Interpret your results.
* The table provides information on the days to maturity
of 40 short-term investments
Days to Maturity Frequency

30–39 3
40–49 1
50–59 8
60–69 10
70–79 7
80–89 7
90–99 4
Estimate the sample mean and sample standard

deviation of the days-to-maturity data.
*Empirical Rule
 The empirical rule states that symmetric or normal distribution with
population mean μ and standard deviation σ have the following
properties.
9
A group of 220 Year 10 students were asked how much time they spent
watching television per week. The results are shown in the table given
below. The mean and standard deviation of hours spent watching television
by the 220 students was 30.32 and 6.04 respectively.
Hours No. of students
10-14 2
15-19 12
20- 24 23
25- 29 60
30- 34 77
35-39 38
40- 44 8
Assuming the frequency distribution is approximately normal,
calculate the interval within which;
i.68% of all observations would expect occur;

ii.95% of all observations would expect occur;
iii.99.7% of all observations would expect occur.
E.G. ANSWER 99% of all observations
This means that there is about a 99% certainty

that an observation will lie between 12 hours
and 48 hours. That is, a student in the sample
will watch between 12 and 48 hours of
television each week.
*Measure of Shape I
*Like central tendency and variation, shape is an important
feature of a distribution.
*Skewness is an important measure of shape.
It is a lack of symmetry.
*A distribution is symmetric if it looks the same to the left

and right of the mean.
*A distribution is skewed if one of its tails is longer than the

other, caused by outliers. 13
Measure of Shape II
Positive skewness Zero skewness or symmetric Negative skewness
Mode < Median < Mean Mean = Median = Mode Mean < Median < Mode
Pearson’s 1st co-efficient of Skewness Pearson’s 2nd co-efficient of Skewness
mean  mod e 3(mean  median)

S .D. S .D.
14
*Example VI
1.Using the data from Example III, compute the 1st and 2nd
Pearson measure of skewness and sketch the diagram.
15
*Bonus Question
The marks out of 10 for forty students who attempted a SOCI1005 test were
recorded as follows:

9, 10, 7, 8, 9, 6, 5, 9, 4, 7, 0, 7, 2, 7, 8, 5, 4, 3, 10, 7,

3, 7, 8, 6, 9, 7, 4, 2, 3, 9, 4, 3, 7, 5, 5, 2, 7, 9, 7, 1.

* a) Construct a grouped frequency table of the scores using 6 classes.
* b) Using the table, calculate the mean, median and mode.
* c) Calculate the variance, standard deviation and coefficient of variation
* d)How would you interpret these results?
16

Descriptive Statistics - Part 3

Uploaded by

Copyright:

Available Formats

Descriptive Statistics - Part 3

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Descriptive Statistics - Part 3

Uploaded by

Copyright:

Available Formats

* Measures of dispersion express quantitatively the degree of

variation or dispersion of values in a population or in

Description Applicability Advantage Disadvantage

Coefficient Measures variability relative 1. Interval/ratio 1. Ensures fairness in 1. Not

Raw data Ungroup frequency Group frequency Notations

Range Max. value – Min Same NA NA

Age (years) No. of Persons

Calculate the standard

Days to Maturity Frequency

Estimate the sample mean and sample standard

Hours No. of students

i.68% of all observations would expect occur;

This means that there is about a 99% certainty

*Skewness is an important measure of shape.

It is a lack of symmetry.

*A distribution is symmetric if it looks the same to the left

*A distribution is skewed if one of its tails is longer than the

Pearson’s 1st co-efficient of Skewness Pearson’s 2nd co-efficient of Skewness

mean  mod e 3(mean  median)

You might also like