FACULTY OF BUSINESS AND MANAGEMENT

BBA116: BASIC STATISTICS

DISTANCE LEARNING

WEEKLY SUMMARISED NOTES

WEEK 1:

TOPIC: Meaning and Purpose of Statistics

SUBTOPICS:
 Introduction to statistics
 Stages in statistical Analysis
 Main divisions of statistics
 Importance of statistics in business
 Limitations of Statistics

Introduction
Statistics may be defined as the science of collection, organization,
presentation, analysis and interpretation of numerical data. A modified form
of this definition is given by Professor Yan-Lun-Chou in whose words,
statistics is a method of decision making in the face of uncertainty on the
basis of numerical data and calculated risks.

According to the above definitions, there are five stages in a statistical

investigation namely: Collection of data, Organization of data, Presentation of
Data, Analysis of data and interpretation.

Collection of data: Collection of data constitutes the first step in a

statistical investigation. Utmost care must be exercised in collecting data
because they form the foundation of statistical analysis. If data are faulty, the
conclusions drawn can never be reliable. The data may be available from
existing published or unpublished sources or else may be collected by the
investigator himself. The first-hand collection of data is one of the most
difficult and important tasks faced by a statistician.

Page 1
Organization of data: Data collected from published sources are generally
in organized form. However, a large mass of figures that are collected from a
survey frequently needs organization. The first step in organizing a group of
data is editing. The collected data must be edited very carefully so that
omissions, inconsistencies, irrelevant answers and wrong computations in the
returns from a survey may be corrected or adjusted. After the data have
been edited the next step is to classify them. The purpose of classification is
to arrange the data according to some common characteristics possessed by
the items constituting the data. The last step in organization is tabulation.
The purpose of tabulation is to arrange the data in columns and rows so that
there is absolute clarity in the data presented.

Presentation of data: After the data have been collected and organized,
they are ready for presentation. Data presented in an orderly manner
facilitates statistical analysis. Data collected may be presented in graphs and
diagrams.

Analysis of data: The purpose of analyzing data is to dig out information

useful for decision making. Methods used in analyzing the presented data
are numerous, ranging from simple observations of the data to complicated,
sophisticated and highly mathematical techniques.

Interpretation: The last in stage in statistical investigation is concerned

with drawing conclusions from the data collected and analyzed. The
interpretation of data is a difficult task and necessitates a high degree of skill
and experience. If the data that have been analyzed are not properly
interpreted, the whole object of the investigation may be defeated and
fallacious conclusions be drawn. Correct interpretation will lead to a valid
conclusion of the study and thus can aid one in making suitable business
decisions. Since statistical methods help in taking decisions, statistics may be
rightly regarded as a body of methods for making wise decisions in the face
of uncertainty.

From the definitions of statistics, we can highlight the major characteristics of

statistics as follows:
 Statistics are the aggregates of facts. It means a single figure is not
statistics. For example, national income of a country for a single year is
not statistics but the same for two or more years is statistics.
 Statistics are affected by a number of factors. For example, sale of a
product depends on a number of factors such as its price, quality,
competition, the income of the consumers, and so on.
 Statistics must be reasonably accurate. Wrong figures, if analysed, will
lead to erroneous conclusions. Hence, it is necessary that conclusions
must be based on accurate figures.
 Statistics must be collected in a systematic manner. If data are collected
in a haphazard manner, they will not be reliable and will lead to
misleading conclusions.
 Statistics is collected in a systematic manner for a pre-determined
purpose.

Page 2
 Statistics should be placed in relation to each other. If one collects data
unrelated to each other, then such data will be confusing and will not lead
to any logical conclusions. Data should be comparable over time and over
space.

Main Divisions of Statistics

Statistics as a science can be divided into two main classes: Statistical
methods and applied statistics.

Statistical Methods: In statistical methods, we study all those devices,

rules of procedures and general principles which are applicable to all kinds or
groups of data. They are the tools in the hands of a statistical investigator.
They are the devices for achieving the desired ends explained in theory.

Applied Statistics: Applied statistics deals with the application of statistical

methods to specific problems. Just like statistical methods, applied statistics
can be further divided into two main groups. These are: Descriptive statistics
and inferential statistics.
 Descriptive Statistics: These are methods involving the collection,
presentation and characterization of data in order to properly describe the
various features of that set of data. Areas covered under descriptive
statistics include: Collection of data, Organization and presentation of
data, Measures of central tendency and Measures of dispersion.
 Inferential statistics: They are those methods that make possible the
estimation of a characteristic of a population or the making of a decision
concerning a population based only on sample results. In other words, it is
the process of reaching generalizations about the whole population by
examining a portion. Areas covered under inferential statistics are:
Probability theory, Sampling distributions, Theory of estimation,
Hypothesis testing.

Importance of Statistics in Business

There are three major functions in any business enterprise in which the
statistical methods are useful. These are as follows:
 The planning of operations: This may relate to either special projects
or to the recurring activities of a firm over a specified period.
 The setting up of standards: This may relate to the size of
employment, volume of sales, fixation of quality norms for the
manufactured product, norms for the daily output, and so forth.
 The function of control: This involves comparison of actual production
achieved against the norm or target set earlier. In case the production has
fallen short of the target, it gives remedial measures so that such a
deficiency does not occur again.

Although these three functions - planning of operations, setting standards,

and control-are separate, in practice they are very much interrelated.

Limitations of Statistics
Statistics has a number of limitations, pertinent among them are as follows:

Page 3
 There are certain phenomena or concepts where statistics cannot be
used. This is because these phenomena or concepts are not amenable to
measurement. For example, beauty, intelligence, courage cannot be
quantified. Statistics has no place in all such cases where quantification is
not possible.
 Statistics reveal the average behaviour, the normal or the general trend.
An application of the 'average' concept if applied to an individual or a
particular situation may lead to a wrong conclusion and sometimes may
be disastrous. For example, one may be misguided when told that the
average depth of a river from one bank to the other is four feet, when
there may be some points in between where its depth is far more than
four feet. On this understanding, one may enter those points having
greater depth, which may be hazardous.
 Since statistics are collected for a particular purpose, such data may not
be relevant or useful in other situations or cases. For example, secondary
data (i.e., data originally collected by someone else) may not be useful for
the other person.
 Statistics are not 100 per cent precise as is Mathematics or Accountancy.
 In statistical surveys, sampling is generally used as it is not physically
possible to cover all the units or elements comprising the universe. The
results may not be appropriate as far as the universe is concerned.
Moreover, different surveys based on the same size of sample but
different sample units may yield different results.
 At times, association or relationship between two or more variables is
studied in statistics, but such a relationship does not indicate cause and
effect' relationship. It simply shows the similarity or dissimilarity in the
movement of the two variables. In such cases, it is the user who has to
interpret the results carefully, pointing out the type of relationship
obtained.
 Statistics does not reveal all pertaining to a certain phenomenon. There is
some background information that statistics does not cover. Similarly,
there are some other aspects related to the problem on hand, which are
also not covered. The user of Statistics has to be well informed and should
interpret Statistics keeping in mind all other aspects having relevance on
the given problem.

Misuse of Statistics
Many people, knowingly or unknowingly, use statistical data in wrong
manner. The misuse of Statistics may take several forms some of which are
explained below:
 Sources of data not given: At times, the source of data is not given. In
the absence of the source, the reader does not know how far the data are
reliable. Further, if he wants to refer to the original source, he is unable to
do so.
 Defective data: Another misuse is that sometimes one gives defective
data. This may be done knowingly in order to defend one's position or to
prove a particular point. This apart, the definition used to denote a certain
phenomenon may be defective. For example, in case of data relating to
unemployed persons, the definition may include even those who are

Page 4
 employed, though partially. The question here is how far it is justified to
include partially employed persons amongst unemployed ones.
 Unrepresentative sample: In statistics, several times one has to
conduct a survey, which necessitates to choose a sample from the given
population or universe. The sample may turn out to be unrepresentative
of the universe. One may choose a sample just on the basis of
convenience. He may collect the desired information from either his
friends or nearby respondents in his neighbourhood even though such
respondents do not constitute a representative sample.
 Inadequate sample: A sample that is unrepresentative of the universe is
a major misuse of statistics. This apart, at times one may conduct a
survey based on an extremely inadequate sample. For example, in a city
we may find that there are 1,000,000 households. When we have to
conduct a household survey, we may take a sample of merely 100
households comprising only 0.1 per cent of the universe. A survey based
on such a small sample may not yield right information.
 Unfair Comparisons: An important misuse of statistics is making unfair
comparisons from the data collected. For instance, one may construct an
index of production choosing the base year where the production was
much less. Then he may compare the subsequent year's production from
this low base. Such a comparison will undoubtedly give a rosy picture of
the production though in reality it is not so. Another source of unfair
comparisons could be when one makes absolute comparisons instead of
relative ones. An absolute comparison of two figures, say, of production or
export, may show a good increase, but in relative terms it may turn out to
be very negligible. Another example of unfair comparison is when the
population in two cities is different, but a comparison of overall death
rates and deaths by a particular disease is attempted. Such a comparison
is wrong. Likewise, when data are not properly classified or when changes
in the composition of population in the two years are not taken into
consideration, comparisons of such data would be unfair as they would
lead to misleading conclusions.
 Unwanted conclusions: Another misuse of statistics may be on account
of unwarranted conclusions. This may be as a result of making false
assumptions. For example, while making projections of population in the
next five years, one may assume a lower rate of growth though the past
two years indicate otherwise. Sometimes one may not be sure about the
changes in business environment in the near future. In such a case, one
may use an assumption that may turn out to be wrong. Another source of
unwarranted conclusion may be the use of wrong average. Suppose in a
series there are extreme values, one is too high while the other is too low,
such as 800 and 50. The use of an arithmetic average in such a case may
give a wrong idea. Instead, harmonic mean would be proper in such a
case.
 Confusion of correlation and causation: In statistics, several times
one has to examine the relationship between two variables. A close
relationship between the two variables may not establish a cause-and-
effect-relationship in the sense that one variable is the cause and the
other is the effect. It should be taken as something that measures degree
of association rather than try to find out causal relationship

Page 5
Key terms in Statistics
A statistic:- is a summary measure that is computed to describe a
characteristic from only a sample of the population. Thus, one major aspect
of inferential statistics is the process of using sample statistics to draw
conclusions about the true population parameters. The need for inferential
methods derives from the need for sampling. As the population becomes
larger, it is usually too costly, too time consuming, and too cumbersome to
obtain our information from the entire population. So, decisions pertaining to
populations characteristics have to be based on the information contained in
a sample of that population.
A parameter:- is a summary measure that is computed to describe a
characteristic of an entire population.
A population (or universe):- is the totality of all things under consideration.
A sample:- is the portion of the population that is selected for analysis.

WEEK 2:

TOPIC: Types of Data and Data Sources

SUBTOPICS:

Page 6
  Statistical data
 Quantitative and Qualitative data
 Primary and Secondary data

Statistical data
Statistical data are the basic raw material of statistics. Data may relate to an
activity of interest, a phenomenon, or a problem situation under study. They
derive as a result of the process of measuring, counting and/or observing.
Statistical data, therefore, refer to those aspects of a problem situation that
can be measured, quantified, counted, or classified. Any object subject
phenomenon, or activity that generates data through this process is termed
as a variable. In other words, a variable is one that shows a degree of
variability when successive measurements are recorded.

In statistics, data are classified into two broad categories: quantitative data
and qualitative data. This classification is based on the kind of characteristics
that are measured.

Quantitative data are those that can be quantified in definite units of

measurement. These refer to characteristics whose successive
measurements yield quantifiable observations. Depending on the nature of
the variable observed for measurement, quantitative data can be further
categorized as continuous and discrete data. A variable may also thus be a
continuous variable or a discrete variable.

 Continuous data represent the numerical values of a continuous

variable. A continuous variable is the one that can assume any value
between any two points on a line segment, thus representing an interval
of values. The values are quite precise and close to each other, yet
distinguishably different. All characteristics such as weight, length, height,
thickness, velocity, temperature, tensile strength, etc., represent
continuous variables. Thus, the data recorded on these and similar other
characteristics are called continuous data. It may be noted that a
continuous variable assumes the finest unit of measurement. Finest in the
sense that it enables measurements to the maximum degree of precision.
 Discrete data are the values assumed by a discrete variable. A discrete
variable is the one whose outcomes are measured in fixed numbers. Such
data are essentially count data. These are derived from a process of
counting, such as the number of items possessing or not possessing a
certain characteristic. The number of customers visiting a departmental
store every day, the incoming flights at an airport, and the defective items
in a consignment received for sale, are all examples of discrete data.

Qualitative data refer to qualitative characteristics of a subject or an object.

A characteristic is qualitative in nature when its observations are defined and
noted in terms of the presence or absence of a certain attribute in discrete
numbers. These data are further classified as nominal and rank data.

Page 7
 Nominal data are the outcome of classification into two or more
categories of items or units comprising a sample or a population according
to some quality characteristic. Classification of students according to sex
(as males and females), of workers according to skill (as skilled, semi-
skilled, and unskilled), and of employees according to the level of
education (as matriculates, undergraduates, and post-graduates), all
result into nominal data. Given any such basis of classification, it is always
possible to assign each item to a particular class and make a summation
of items belonging to each class. The count data so obtained are called
nominal data.
 Rank data are the result of assigning ranks to specify order in terms of
the integers 1,2,3, ..., n. Ranks may be assigned according to the level of
performance in a test, a contest, a competition, an interview, or a show.
The candidates appearing in an interview, for example, may be assigned
ranks in integers ranging from I to n, depending on their performance in
the interview. Ranks so assigned can be viewed as the continuous values
of a variable involving performance as the quality characteristic.

Data sources could be seen as of two types, viz., secondary and primary. The
two can be defined as under:
 Secondary data: They already exist in some form: published or
unpublished - in an identifiable secondary source. They are, generally,
available from published source(s), though not necessarily in the form
actually required.
 Primary data: Those data which do not already exist in any form, and
thus have to be collected for the first time from the primary source(s). By
their very nature, these data require fresh and first-time collection
covering the whole population or a sample drawn from it.

Methods of collecting primary data

 Personal interviews
 Telephone interviews
 Mail interview

Sources of secondary data

Chief sources of secondary data may be broadly classified into the following
two groups:
 Published sources
 Unpublished sources
Published sources
Include publications of various organisations such as;
i. Official publications of central government – i.e., economic surveys
ii. Annual and monthly abstracts of statistics- i.e statistical abstracts
iii. Publications of research institutions
iv. Publications of commercial and financial institutions.
v. Newspapers and periodicals.
vi. International publications e.g. U.N.O, WHO, IMF

Unpublished sources
There are various sources of unpublished statistical material such as;

Page 8
 Records maintained by private firms or business enterprises who may not
want to release their data to any outside agency.
 The various departments and offices of the government.
 The researches carried out by the individual research scholars in the
universities or research institutes

Precautions in the use of secondary data

 Secondary data should be used with extra caution.
 Before using such data, the investigator must be satisfied regarding the
reliability, accuracy, adequacy and suitability of the data to the given
problem under investigation
 Proper care should be taken to edit it so that it is free from
inconsistencies, errors and omissions

WEEK 3:

TOPIC: Data Display and Presentation

SUBTOPICS:
 Diagrammatic and Graphic presentation of data
 Classification of data
 Frequency distribution

Classification and tabulation of data are devices of presenting statistical data

in neat, concise, systematic and readily comprehensible and intelligible form,
thus highlighting the salient features. Data can also be presented in diagrams
and graphs.

Diagrammatic and graphic presentation of statistical data

Construction of a good graph or diagram requires the following:
i. Clear, concise and unambiguous titles.
ii. Clear and concise statement of the units in which the figures are to be
measured
iii. Correct vertical and horizontal scaling
iv. Statement of units used in the vertical and horizontal scale.
v. Graph or diagram should be tidy and attractive.
vi. Statement of the source of information, at the bottom.
vii. A key to explain the various features of a graph

Bar Diagrams
 Bar diagrams are one of the easiest and the most commonly used devices
for presenting most of economic and business data.
 Consist of a group of equidistant rectangles, one for each group or
category of the data in which the values or the magnitude are represented
by the height or length of the rectangle, width being arbitrary and
immaterial.

Points to consider when drawing bar diagrams

Page 9
 All the bars drawn in a single study should be of uniform (though
arbitrary) width depending on the number of bars to be drawn and the
space available.
 Proper but uniform spacing should be given between different bars to
make the diagram look more attractive and elegant.
 The height (length) of the rectangles or bars are taken proportional to
magnitude of observations.
 The scale being selected keeping in view the magnitude of the largest
observation.
 All bars should be constructed on the same baseline
 Where possible, bars should be arranged from left to right.

Types of bar diagrams

 Simple bar diagram
 Component( or sectional) bar diagram
 Percentage bar diagram
 Multiple bar diagram
 Bilateral bar diagram

Graphic Representation of data

Like diagrams, a large number of graphs are used in practice. But they can be
broadly classified under the following two heads.
 Graphs of frequency distributions
 Graphs of time series

Graphs of frequency Distributions

 These are designed to reveal clearly the characteristics of a frequency
data
 Such graphs are more appealing to the eye than the tabulated data and
are readily perceptible to the mind
 They facilitate comparative study of two or more frequency distributions
regarding their shape and pattern

Examples of graphs of frequency distributions

 Histogram
 Frequency polygons
 Frequency curves
 Ogives or cumulative frequency curves

Classification of data
Classification is the grouping of related facts into classes. Facts in one class
differ from those of classification.

Objectives of Classification
 To condense the mass of data in such a manner that similarities and
dissimilarities can be readily apprehended.
 To facilitate comparison
 To pinpoint the most significant features of the data at a glance.

Page 10
  To give prominence to the information gathered while dropping out the
unnecessary elements
 To enable a statistical treatment of the material collected.

Types of Classification
Data can be classified broadly on the basis of the following 4 criteria:
1. Geographical i.e. area-wise e.g. cities, districts e.t.c
2. Chronological i.e. on the basis of time
3. Qualitative i.e. according to some attributes
4. Quantitative i.e. in terms of magnitude
Geographical Classification: Data are classified on the basis of
geographical or vocational differences between the various items like states,
cities, regions, zones, areas e.t.c.

Chronological Classification: When data are observed over a period of

time the type of classification is known as chronological classification e.g. we
may present the figures of population (or production, sales e.t.c) as follows:

Population of the country from 1999 to 2017(hypothetical figures)

Year Pop. In Millions
1999 14 million
2008 21 million
2017 28 million
Time series are usually listed in chronological order; normally starting with
earliest period.

Qualitative Classification: Data are classified on the basis of some

attribute or quality such as sex, colour of hair, literacy, religion e.t.c. The
attribute under study cannot be measured; one can only find out whether it is
present or absent in the units of the population under study. E.g. if the
attribute under study is population, one can find out how many persons are
living in urban areas and how many in rural areas. Thus, when only one
attribute is studied, one having the attribute and the other not possessing the
attribute. This type of classification is called simple classification.

Quantitative Classification: Quantitative classification refers to

classification of data according to some characteristic that can be measured
such as height, weight, income, sales, profits, production e.t.c., e.g. the
students in college may be classified according to weight as follows:

Weight in No. of
(lb) Students
90-100 50
100-110 200
110-120 260
120-130 360
130-140 90
140-150 40

Page 11
 Total 1000
Such a distribution is known as empirical frequency distribution or simple
frequency distribution.

A frequency distribution refers to data classified on the basis of some

variable that can be measured such as prices, wages, age, number of units
consumed or produced.
A frequency distribution or frequency table is simply a table in which the
data grouped into classes and the numbers of cases which fall in each class
are recorded. The numbers in each class are referred to as ‘frequencies’,
hence the term ‘frequency’. When the number of items are expressed by
their proportion in each class, the table is usually referred to as a ‘relative’
frequency distribution’, or simply a percentage distribution.

Formation of a Discrete Frequency Distribution

We count the number of times a particular value is repeated which is called
the frequency of that class. To facilitate counting we prepare a ‘tally’ column.

Example:
In a survey of 35 families in a village, the number of children per family was
recorded and the following data obtained:
1 0 2 3 4 5 6
7 2 3 4 0 2 5
8 4 5 12 6 3 2
7 6 5 3 3 7 8
9 7 9 4 5 4 3
Represent the data in the form of a discrete frequency distribution.

Solution:
Frequency distribution of the number of children

No. of children Tallies Frequency

1 II 2
2 I 1
3 IIII 4
4 IIII 5
5 III 3
6 IIII 5
7 IIII 4
8 II 2
9 II 2
10 - 0
11 - 0
12 I 1

Formation of a Continuous Frequency Distribution

Class Intervals and Class limits are important terms when a continuous
frequency distribution is formed.

Page 12
Class limits: These are the highest and lowest values that can be included in
the class. For example take the class 20-40. The lowest value of the class is
20 and the highest is 40. The two boundaries of class are known as the lower
limit and the upper limit of the class.
Class intervals: The difference between the upper and lower limit of a class
is known as class interval of that class. For example, in the class 100-200, the
class interval is 100 (i.e. 200-100).

An important decision while constructing a frequency distribution is about the

width of the class interval, i.e. whether it should be 10, 20, 50, 100, 500 e.t.c.
The decision would depend upon a number of factors such as the range in the
data, i.e. the difference between the smallest and the largest item, the
details required and the number of classes to be formed e.t.c. A simple
formula to obtain the estimate of appropriate class interval, i.e. i is:
L−S
i=
K
Where: L – Largest item
S – Smallest item
K – The number of classes

Example
If the salary of 100 employees in a commercial undertaking varied between
£500 and £5, 500, and we want to form 10 classes, then the class interval
would be:
L−S
i=
K
L = 5, 500
S = 500
K = 10
5 ,500−500
i=
10
= 500
The starting class would be 500-1000, the next 1000 – 1500 and so on.

The question now is how to fix the number of classes, i.e. K. The number can
be either fixed arbitrarily keeping in view the nature of problem under study
or it can be decided with the help of Sturge’s rule. According to him, the
number of classes can be determined by the formula:
K = 1 + 3.322 log N
Where: N = total number of observations
Log = logarithms of the number
Thus if 10 observations are being studied, the number of classes shall be:
K = 1 + (3.222 x 1)
= 4.322 or 4

And if 100 observations are being studied, the number of classes shall
generally be between 4 and 20 – it cannot be less than 4 even if N is less
than 10 and if N is 10 K will be 1 + 3.222 x 6 = 20.9 or 21

Page 13
Sturges suggested the following formula for determining the magnitude of
class interval:

Range
i=
1+3 .322 log N
Where range is the difference between the largest and smallest items. e.g. in
the example above; and by applying the formula, the magnitude of class
interval shall be:

5500−500
i=
1+3 .22 log 100
5000
= 7 .644
= 654.1 or 650

Class Frequency
The number of observations corresponding to a particular class is known as
the frequency of that class or the class frequency.

Class Midpoint or class mark

It is the value lying halfway between the lower and upper class limits of a
class interval. The midpoint of a class is ascertained as follows:

Midpoint =
There are two methods of classifying the data according to class-intervals:
i) Exclusive method
ii) Inclusive method

Exclusive Method
The class intervals are so fixed that the upper limit of one class is the lower
limit of the next class.

Example
Income No. Of Persons
1000-1100 50
1100-1200 100
1200-1300 200
1300-1400 150
1400-1500 40
1500-1600 10
Total 550

Page 14
Inclusive Method
The upper limit of one class is included in that class itself.

Example:
Income (£) No. of Persons
1000 – 1099 50
1100 – 1199 100
1199 – 1299 200
1300 – 1399 150
1400 – 1499 40
1500 – 1599 10
Total 550

To decide whether to use the inclusive or exclusive method, it is important to

determine whether the variable under observation is a continuous or discrete
one. In case of continuous variables, the upper limit exclusive method must
be used. e.g. the variable height being inherently a continuous one should
be stated as 60” and under 62”, and under 64 and so on. The inclusive
method should, in general be used in case of discrete variables.

Relative Frequency Distribution

They show the percentage of the total number of observations in each class.
E.g. it is obtained by dividing each of the class frequencies by the total
number of frequencies so that the relative frequencies will total to one.
Consider the example above.

WEEK 4:

TOPIC: Measures of Central Tendency

SUBTOPICS
 Mean
 Arithmetic Mean
 Geometric Mean
 Harmonic Mean

Few people can assimilate a mass of detailed information expressed in

numerical form, even when it has been substantially reduced by tabulation.
It is helpful, therefore, if instead of merely tabulating the information derived
from a specific enquiry and depicting it in graphs and diagrams, it can be
expressed in a more abbreviated numerical form, yet in such a way that the
salient features of the tables are clearly brought out.

For the purpose of statistical decision it is essential to extract from the data
just a few numbers, which summarize the essential information. These
summarized numbers are called:
(i) Statistics if the information is from a sample and

Page 15
(ii) Parameters if the information is from the whole Population, where
here by population we mean the totality of all possible observations.

Types of averages
- Arithmetic mean
- Geometric mean
- Harmonic mean
- Median
- Mode

Arithmetic Mean
i) For discrete data/ungrouped data:

x=
∑ xi
n
Where: Xi = individual observations
N = Number of observations
For Example, given the data below find the mean:
2 3 4 5 6

x=
∑ xi
n
2+ 3+4 +5+6
= 5
ii) For grouped data/continuous data
Σ fx
x=
Σx

Where x = mean
f =frequencies for each class
x = midpoints of the class intervals
Illustration
For example, given the data below; compute Arithmetic Mean
Marks 0-10 10-20 20-30 30-40 40-50 50-60

No. of 5 10 25 30 20 10
Students

Solution
Marks Midpoint (x) F Fx

0-10 5 5 25

Page 16
10-20 15 10 150

20-30 25 25 625

30-40 35 30 1050

40-50 45 20 900

50-60 55 10 550

∑f =100 ∑fx = 3300

Σ fx 3300
x=
Σx =100 = 33
Another method:
Σ fd
x= A+
Σf

Where: A = assumed mean

d = deviations of midpoints from assumed mean
i.e. d = x – A
∑f = total frequencies
We apply the method to the data above.
We take 35 as assumed mean

Mark Midpoint f d(x – fd

(x) 35)

0-10 5 5 -30 -150

10-20 15 10 -20 -200

20-30 25 25 -10 0

30-40 35 30 0 0

40-50 45 20 +10 +200

50-60 55 10 +20 +200

∑f =100 ∑fd = -
200

Σ fd 200
x= A+
Σf = 35 - 100 = 33

Page 17
Advantages of Arithmetic Mean
- It can be easily understood.
- It takes into account all the items of the series.
- It is not necessary to arrange the data first and then calculate
arithmetic average.
- It is capable of algebraic treatment.
- It is a good method of comparison
- It is used frequently

Disadvantages
- It is affected by extreme values to a great extent.
- It may be a figure which does not exist in a series.
- It cannot be calculated if all the items of a series are not known.
- It cannot be used in case of qualitative data.

Properties of Arithmetic Mean

1. The sum of the deviations of the items from arithmetic mean, taking
signs into account is always zero. i.e. Σ( x−x)=0 . This would be clear
from the following example.

x (x - x )
10 -20
20 -10
30 0
40 +10
50 +20_______
∑x = 150 Σ( x−x)=0
150
Here x = 5 = 30

Due to this property, mean is characterized as point of balance i.e. the sum of
the positive deviations from it is equal to the sum of the negative deviations
from it.

2. The sum of the squared deviations of items from arithmetic mean is

minimum i.e. less than the sum of the squared deviations of the items from
any other value.

Page 18
 Σ xi
x=
3. Since N or N x=Σx in other words, if we replace each item
in the series by the mean, then the sum of these substitutions will be equal to
the sum of the individual items.

4. If we have the arithmetic mean and number of items of two or more

than two related groups, we can compute combined average of these groups
by applying the following formula.
N 1 x 1+ N 2 x 2
x 12=
N 1+ N 2

Where; x 12= Combined mean of the two groups

x 1 = Arithmetic mean of first group
x 2 = Arithmetic mean of second group
N1 = Number of items in the first group
N2= Number of items in the second group
Illustration
The mean height of 25 male workers in a factory is 61 inches and the mean
height of 35 female workers in the same factory is 58 inches. Find the
combined mean height of 60 workers in the factory.

N 1 x 1+ N 2 x 2
x 12=
N 1+ N 2
N1 = 25

x 1 = 61
N2 = 35

x2 = 58
(25×61 )+(35×58)
x 12=
25+35
= 59.25

Geometric Mean
Is defined as the Nth root of the product of N items or values. If there are two
items, we take the square root; if there are three items, the cube root; and so
on. Symbolically:

GM =
N √( x 1 )×( x 2 )×( x 3 )×. ... .
Where x1, x2, x3, ……e.t.c. refer to the various items of the series.

Page 19
For example, compute GM for the data below:
2, 3, 4

GM = √3 x 2∗x 2∗x 3
= √3 2×3×4
= √3 24
= 2.88

When the number of items is three or more the task of multiplying the
numbers and extracting the root becomes excessively difficult. To simplify
calculations logarithms are used.
GM is then computed as follows:
Σ log x
GM = antilog N for ungrouped data.
For grouped data:
Σf log x
GM = antilog Σf
Where: x = midpoint of class intervals
f = frequencies of each class

Illustration
Daily income of ten families of a particular place is given below. Find out the
Geometric Mean.
85 70 15 75 500 8 45 250 40 36
Solution
X log x
85 1.9294
70 1.8451
15 1.1761
75 1.8751
500 2.6990
8 0.9031
45 1.6532
250 2.3979
40 1.6021
36 1.5563
∑logx= 17.6373

Σ log x
GM = antilog N

Page 20
GM = antilog
=antilog 1.76373
=58.08

Illustration
Find the geometric mean for the data below:
Marks Frequency x (midpoint) Log x f log x
4-8 6 6 0.7782 4.6692
8-12 10 10 1.000 10.000
12-16 18 14 1.1461 20.5303
16-20 30 18 1.2553 37.6590
20-24 15 22 1.3424 20.1360
24-28 12 26 1.4150 16.9800
28-32 10 30 1.4771 14.7710
32-36 6 34 1.5315 9.1890
36-40 2 38 1.5798 3.1596
∑f=109 ∑flogx=137.1936

Σf log x
GM = antilog Σf
137 . 1936
= antilog 109
= antilog 1.2587
= 18.14
Uses of Geometric Mean
GM is specifically useful in the following cases:
a) Gm is used to find the average percent increase in sales, production,
population or other economic or business series. E.g. from 1996-1998,
prices increased by 6, 10 & 18 percent respectively. The average
5+10+18
=11
annual increase is not 11 percent ( 3 ) as given by
arithmetic mean but 9.65 percent as obtained by the GM.
b) GM is theoretically considered to be the best average in the
construction of index numbers. It satisfies the time reversal test and
gives equal weight to equal ratio change.
c) It is most suitable when large weights have to be given to small items
and small weights to large items, situations which we usually come
across in social and economic fields.

Harmonic Mean

Page 21
Is mean based on the reciprocals of numbers averaged. It is defined as the
reciprocal of arithmetic mean of the reciprocal of the individual observation.
Thus for ungrouped data:
N
1
Σ
HM = xi

For grouped data:

Σf
Σf xi1
HM =
Illustration
Find the harmonic mean from the following:
2574 475 75 5 0.8 0.08 0.005 0.0009

x 1
x

2574 0.0004

475 0.0021

85 0.0133

5 0.2000

0.8 1.2500

0.08 12.5000

0.005 200.0000

0.0009 1111.1111

Σ 1x
=1325.0769

N=8
N
1
Σ
Therefore: HM = x

Page 22
 8
= 1325 . 0769
= 0.006

Illustration
Compute HM from the following data
Class f x(midpoint) 1
interval
fx
10-20 4 15 0.267
20-30 6 25 0.240
30-40 10 35 0.286
40-50 7 45 0.156
50-60 3 55 0.155
∑f=30 Σf 1x
= 1.004

Σf

HM =
Σf
1
()
x
30
= 1. 004
= 29.88

Uses of Harmonic Mean

Harmonic Mean is restricted in its field of usefulness. It is useful for
computing the average rate of increase in profits of a concern or average
speed at which a journey has been performed or the average price at which
an article has been sold. The rate usually indicates the relation between two
different types of measuring units that can be expressed reciprocally. E.g. if
a man walked 20km in 5 hrs, the rate of his walking speed can be expressed
as:
20 Km
5 hrs = 4 km per hour
Where the unit of the first term is a km and the unit of the second term is an
hour or reciprocally

Hrs per km

WEEK 5:

TOPIC: Measures of Central Tendency

SUBTOPICS

Page 23
  Median
 Quartiles and Percentiles
 Mode

Median
Refers to the middle value in a distribution when these values are arranged in
n+1
ascending or descending order. The formula for finding median item is 2
Where n = number of items. When the half value is required then median is
more suitable.

Median is calculated as under:

Median for discrete data/ungrouped data

n+1
If the number of items is odd, the formulae will be 2 th item.
Arranging the items in ascending order,

Example
Compute median from the following data
2 8 5 9 3
Solution

= 3rd item

Arranging in ascending order we obtain

2 3 5 8 9
Now, since the third item is 5, it follows that

When the number of data items is even, median is obtain by the formula

Median =

For Example
Obtain median from the data below
112, 115, 140, 110, 108, 116

Solution

Page 24
So the following formula will apply

Median =
We now arrange the data in ascending order to obtain the following series

108, 110, 112, 115, 116, 140

1 6
Median = 2 2
[ thitem+
6+ 2
2
thitem
]
1
[ 3 rditem+4 thitem ]
= 2
1
[ 112+115 ]
= 2
= ½ x 227
= 113.5

Median for continuous series

N
Determine the particular class in which the value of median lies. Use 2 as
( N +1)
the rank of the median and not 2 . Some writers have suggested that
while calculating median in continuous series, 1 should be added to total
frequency if it is odd (say 99) and should not be added if it is an even figure
(say 100). However, 1 is to be added in case of individual and discrete series
because specific items and individual values are involved. In continuous
frequency distribution all the frequencies lose their individuality. The effort
now is not to find the value of one specific item but to find a particular point
on a curve – that value which will have 50% of frequencies on one side of it
and 50% of the frequencies on the other. Hence it will be wrong to use the
N
above rule. Hence it is 2 which will divide the area of curve into two equal
( N +1)
parts and as such we should use 2 in continuous series.

The formula is:

i
L+ (m−c )
Median = f
L= lower class limit of the median class i.e. the class in which the middle
item of the distribution lies.
N
m = 2 th item, n being the total frequencies
i = class interval of the median class

Page 25
f = frequency of the median class
c= cumulative frequencies of the class preceding the median class

Illustration
Compute the median for the following frequency distribution:
Marks No. of Students
45-50 10
40-45 15
35-40 26
30-35 30
25-30 42
20-25 31
15-20 24
10-15 15
5-10 7

Solution
Arrange the data in ascending order and then find out the median.
Calculation of median
Marks F cf
5-10 7 7
10-15 15 22
15-20 24 46
20-25 31 77
25-30 42 119
30-35 30 149
35-40 26 175
40-45 15 190
45-50 10 200

N = 200
N
M= 2
200
= 2
= 100th item

Median lies in the 25-30 class

L = 25
i = 30-25 = 5
f = 42
c = 77
i
L+ (m−c )
Median = f
5
25+ (100−77 )
= 42
= 27.74

Page 26
Illustration
Compute median from the data below:
Weight in gms No. of Apples
410-419 14
420-429 20
430-439 42
440-449 54
450-459 45
460-469 18
470-479 7

Solution
Since we are given inclusive class intervals we should convert them to
exclusive one by deduction 0.5 from the lower limit and adding 0.5 to the
upper limits.
Weight f cf
409.5 - 419.5 14 14
419.5 – 429.5 20 34
429.5 - 439.5 42 76
439.5 – 449.5 54 130
449.5 - 459.5 45 175
459.5 – 469.5 18 193
469.5 – 479.5 7____ 200
N = 200
N
M=2
200
= 2
= 100
Median lies in the 439.5 – 449.5 class
L = 439.5
i = 10
f = 54
c = 76
i
L+ (m−c )
median = f
10
439 . 5+ (100−76 )
= 54
= 443.9

Illustration
Find median from the data below:
Marks No. of Students cf
0-5 29 29
5-10 195 224
10-15 241 465

Page 27
15-20 117 582
20-25 52 634
25-30 10 644
30-35 6 650
35-40 3 653
40-45 2 655
N
Median = size of 2 th item
N 655
M= 2 = 2 = 327.5th item
Median lies in the 10-15 class
M = 10
i=5
f = 241
c = 224
i
L+ (m−c )
median = f

5
10+ (327 . 5−224 )
= 241
= 12.15

Calculating median when class intervals are unequal

When the class intervals are unequal, the frequencies need not to be
adjusted to make the class intervals equal and the same formula for
interpolation can be applied as discussed above.

Illustration
Calculate the median from the following data

Marks 0-10 10-30 30-60 60-80 80-90

No. of students 5 15 30 8 2
Solution
Since class intervals are unequal, let us first convert it to a distribution with
equal class intervals on the assumption that the frequencies are equally
distributed throughout a class.
Calculation of Median

Marks f cf
0-10 5 5
10-30 15 20
30-60 30 50
60-80 8 58
80-9- 2 60

N = 60

Page 28
 N
M= 2
60
2 = 30th item

Median class is 30-60

i
L+ (m−c )
Median = f
30
30+ (30−20 )
= 30
= 40
Merits of Median
- It is especially useful in case of open-end classes since only the
position and not the values of the items must be known. It is also
recommended if the distribution has unequal classes, since it is easier
to compute than the mean.
- Extreme values do not affect the median as strongly as they do the
mean. E.g. the median of 10, 20, 30, 40 and 150 is 30. Hence the
mean is 50. Hence very often when extreme values are present in a
set of observations, the median is a more satisfactory measure of the
central tendency than the mean.
- In markedly skewed distributions such as income distributions or price
distributions where the arithmetic mean would be distorted by extreme
values, the median is especially useful. Consequently, the median
income for some purposes can be regarded as a more representative
figure, for half the income earners must be receiving at least the
median income.
- It is the most appropriate average in dealing with qualitative data i.e.
where the ranks are given or there are other types of items that are
not counted or measured but are scored.
- Value of median can be determined graphically whereas the value of
mean cannot be graphically ascertained.
- It actually indicates what many people incorrectly believe the
arithmetic mean indicates. The median indicates the value of the
middle item in the distribution. This is a clear cut meaning and makes
the median a measure that can be easily explained.

Demerits of Median
- To compute the median, it is necessary to arrange the data: other
averages do not need any arrangement.
- Since it is a positional average, its value is not determined by each and
every observation.
- It is not capable of algebraic treatment e.g. median cannot be used to
determine the combined median of two or more groups as is possible
in the case of mean.
- The value of median is affected more by sampling fluctuations than the
value of arithmetic mean.

Page 29
 - The median, in some cases cannot be computed exactly as the mean.
When the number of items included in a series of data is even, the
median is determined approximately as the midpoint of the two middle
items.
- It is erratic if the number of items is small.
Usefulness:
- The median is useful for distributions containing open ended intervals
since these intervals do not enter its computation. Also since the
median is affected by the number rather than the size of items, it’s
frequently used instead of the mean as a measure of central tendency
in cases where such values are likely to distort the mean.

Related Positional Measures

- Besides median, there are other measures which divide a series into
equal parts. Important among these are quartiles, deciles and
percentiles. Quartiles are those values of the variate which divide the
total frequency into four equal parts, deciles divide the total frequency
into 10 equal parts and percentiles divide the total frequency into 100
equal parts. Just as one point divides a series into two equal parts,
three points would divide into 4 equal parts, 9 points into 10 equal
parts and 99 points into 100 parts.
- Consequently, there are only 3 quartiles, 9 deciles and 99 percentiles
for a series.
- Quartiles are denoted by Q and deciles by D and percentiles by P. the
subscripts 1, 2, 3, e.t.c beneath Q, D and P denote the value that we
want to compute.
- Graphically any set of these partitions values divides the area of the
frequency curve or histogram into equal parts. If vertical lines are
drawn as third quartiles for example, the area of the histogram will be
divided by these lines into four equal parts. The 9 deciles divide the
area of the histogram or frequency curve into 10 eaual parts and the
99 percentiles divide the area into 100 equal parts.
- In business statistics and economics quartiles are more widely used
than deciles and percentiles. Quartiles are points on the x-scale that
divide the distribution into four equal parts. Obviously, there are three
quartiles, the second coinciding with the median, more precisely, one-
fourth of the total frequency is less than Q 1 and three fourths is greater
than Q1. The upper quartile, Q3 is that point on the x-axis such that 3
4ths of the total frequency is below Q 3 and one fourth is above it.
- Deciles and percentiles are important in psychological and educational
statistics concerning grades, rates, ranks e.t.c they are of use in
economics and business statistics in personnel work, productivity
ratings and other such situations.

Computation of Quartiles, Percentiles and Deciles

The procedure for computing quartiles, deciles, e.t.c is the same as the
median. While computing these values in individual and discrete series we
add 1 to N whereas in continuous series, we do not add 1.

Page 30
 N +1
Thus Q1 = size of 4 th item (for ungrouped data/individual series)
For grouped data,
N
Q1 = size of the 4 th item.
i ( N −C )
L+
Thus Q1 = f 10
N
Where 4 = M
3( N +1)
Q3 = size of 4 th item (in ungrouped data)
For grouped data

Q3 =
L+ (
i 3N
f 4
−C )
4 ( N +1)
D4 = the size of 10 th item for individual series
For grouped data:
4N
D4 = size of 10 th item

D4 =
L+
f (
i 60 N
100
−C )
th item

Example:
Compute the lower and upper quartiles, 3 rddeciles and 20th percentile from
the following data:
Central value : 2.5 7.5 12.5 17.5 22.5
Frequency: 7 18 25 30 20
Solution:
Since we are given midpoints, we will first find the lower and upper limits of
the various classes. The method for finding these limits is to take the
difference between the two central values, divide it by 2, deduct the values
so obtained from the lower limit and add it to the upper limit. In the given
case:
7 .5−2. 5 5
2 =2
The first class shall be -05, second 5-10 e.t.c.
Calculation of Q1, Q3, D3 and P20
Class group f cf
0-5 7 7
5-10 18 25
10-15 25 50
15-20 30 80

Page 31
20-25 20 100

N
Lower quartile i.e. Q1 = size of 4 th item
100
= 4 th item
= 25th item
Q1 lies in 5-10 class
L=5
i = 10-5 = 5
f = 18
c=7

5
5+ (25−7 )
Q1 = 18 = 10
Upper quartile
3N
Q3 = size of 4 th item
3×100
= 4 = 75th item
Q3 lies in the class 15-20

Q3 =
L+ (
i 3N
f 4
−c )
L = 15, f = 30, c = 50, i = 5
75−50
15+ ×5
Q3 = 30 = 19.17
Third decile:
3N
D3 = size of 10 th item
3×100
= 10
= 30th item
D3 lies in the class 10-15
L = 10, i = 5, f = 25, c = 25

D3 =
i 3N
L+
f 10 (
−c )
5
10+ (30−25 )
= 25
= 10 + 1
= 11
th
20 percentile

Page 32
 20 N
P20 = size of 100 th item
20×100
= 100
= 20th item
P20 lies in the 5-10 class
L = 5, I = 5, f = 8, c = 7

P20 =
L+
f (
i 20 N
100
−c )
5
5+ (20−7 )
= 18
= 8.61

Mode
It is that value in a series of observations which occurs with the greatest
frequency.
The mode of a distribution is the value at the point around which the items
tend to be most heavily concentrated. It may be regarded as the most
typical of a series of values.
A set of data may have a single mode in which case it is said to be unimodal,
it may have two modes which makes it bimodal or it may have several modes
and be called multimodal.

Mode for ungrouped data

Mode for ungrouped data is obtained just by inspection. You consider the
most frequent item in the series.
Illustration
Compute the mode from the following data:
Size of shoes 5 6 7 8 9 10 11
No. of persons 10 20 25 40 22 15 6
The modal size is 8 since it appears maximum number of times in the series.

Computing mode for grouped data

There are two popular methods:
a) We can apply the formula for mode i.e.

( )
Δ1
i
Δ1+ Δ2
Mode = L +
where: L= lower class
Modal group = class with the highest frequency
Δ 1 = difference between the highest frequency and the
frequency
of the preceding class
Δ 2 = difference between the highest frequency and the
frequency of the succeeding class.
i= class interval of the modal group.

Page 33
 Another form of this formula is:

( )
f 1 −f 0
L+ i
2 f 1−f 0 −f 2
Mode =
Where: L = lower limit of the modal class
f1 = frequency of the modal class
f0 = frequency of the class preceding modal class
f2 = frequency of the class succeeding the modal class
i= class interval of the modal group
While applying the above formula for computing mode, it is necessary to see
that the class intervals are uniform, throughout. If they are unequal, they
should first be made equal on the assumption that the frequencies are
equally distributed throughout the class, otherwise we will get misleading
results. Where mode is ill defined, its value may be ascertained by the
following formula based on the relationship between mean, median and
mode: Mode = 3median – 2mean

Illustration
Compute the mode from the following data:
Marks No. of Students
Above 0 80
Above 10 77
Above 20 72
Above 30 65
Above 40 55
Above 50 43
Above 60 28
Above 70 16
Above 80 10
Above 90 8
Above 100 0

Solution
Since this is a cumulative frequency distribution, we first convert into a
simple frequency distribution.
Marks No. of students
0-10 3
10-20 5
20-30 7
30-40 10
40-50 12
50-60 15

Page 34
60-70 12
70-80 6
80-90 2
90-100 8
By inspection, the modal group is 50-60
Therefore: L = 50
Δ1 = 3
Δ2 = 3
i = 10

( )
Δ1
i
Δ1+ Δ2
Mode = L +

= 50 +
( 3+33 ) 10
= 55

Illustration
Find the value of mode from the data given below:
Weight (Kg.) No. of Students
93-97 2
98-102 5
103-107 12
108-112 17
113-117 14
118-122 6
123-127 3
128-132 1

By inspection, mode lies in the class 108-112. But the real limits are 107.5 –
112.5

( )
Δ1
i
Δ1+ Δ2
Mode =L +

L = 107.5
Δ1 = 17-12 = 5
Δ2 = 17-14 = 3
i=5

x=107 . 5+ ( )
5
5+3
5
= 110.625
Thus modal weight is 110.625Kg.

Merits of Mode
By definition mode is the most typical or representative value of a
distribution. Hence when we talk of a modal wage, modal size of shoe or
modal size of family, it is this average that we refer to. The mode is a

Page 35
measure which actually indicates what many people incorrectly believe the
arithmetic mean indicates. It is the most frequently occurring value.
- The median, mode is not unduly affected by extreme values. Even if
the high values are very high and the low values are very low we
choose the most frequent value of the data.
- Its value can be determined in open-ended distributions without
ascertaining the class limits.
- It can be used to describe qualitative phenomenon. E.g. if we want to
compare the consumer preferences for different types of products, say
soap, toothpaste, e.t.c or different media of advertising we should
complete the modal preferences expressed by different groups of
people.
- The value of mode can be determined graphically whereas the value of
mean cannot be graphically ascertained.

Limitations of Mode
- The value of mode cannot always be determined. In some cases we
have a bimodal series.
- It is not capable of algebraic manipulations e.g. from the modes of two
sets of data we cannot calculate the overall mode of the combined
data. Similarly the modal wage times the number of workers will not
give the total payroll-except, of course, when the distribution is normal
and then the mean, median and mode are all equal.
- The value of mode is not based on each and every item of the series.
- It is not a rigidly defined measure. There are several formulae for
computing the mode, all of which usually give somewhat different
answers. In fact, mode is the most unstable average .Its value is
difficult to determine.
- While dealing with quantitative data, the disadvantages of the mode
outweigh its good features and hence it is seldom used.

Usefulness
The mode is employed when the most typical value of a distribution is
desired. It is the most meaningful measure of central tendency in case of
highly skewed or non-normal distributions, as it provides the best indication
of the point of maximum concentration.

Relationship among Mean, Median and Mode

A distribution in which the values of mean, median and mode coincide (i.e.
mean=median=mode) is known as a symmetrical distribution.
Conversely stated, when the values of mean, median and mode are not equal
the distribution is known as asymmetrical or skewed.

In moderately skewed or asymmetrical distributions a very important

relationship exists among mean, median and mode.
Mode = mean – 3[mean-median]
Mode = 3 median – 2 mean

Page 36
 2
and Median = mode + 3 [mean-mode]
Illustration
In a moderately asymmetrical distribution, the mode and mean are 32.1 and
35.4 respectively. Find out the value of median.
Solution
Mode = 3 median – 2 mean
32.1 = 3 median – 2 x 35.4
3 median = 32.1 + 2 x 35.4
3 median = 102.9
102 .9
Median = 3
= 34.3
Illustration
The median and mode of the following wage distribution are known to be
£33.5 and 34 respectively. Three frequency values from the table are
however missing. Find these missing values:
Wages in (£) No. of Workers
0-10 4
20-20 16
20-30 ?
30-40 ?
40-50 ?
50-60 6
60-70 4
Total 230
Let the missing frequencies be:
20-30 x
30-40 y
40-50 230-30-x-y
Since median and mode are 33.5 and 34, they both lie in the 30-40 class.
i
L+ (m−c )
Median = f
L = 30
i = 10
f=y
c = 20 + x
10
(115−20−x )
33.5 = 30 + y
33.5y = 30y + 1150 – 200 – 10x
3.5y = 950 – 10x
3.5y + 10x = 950……………….(i)
Mode is 34, so modal class is 30-40
L = 30,
Δ1 = y - x,
Δ2 = y - [230 – 30 – x - y]
= y – 230 + 30 + x + y
= 2y – 200 + x

Page 37
 Mode =

Therefore: 34 =
y− x
30+
= y− x +2 y−200+ x

34×
1
10
1
=30× + (
y−x
10 3 y−200
10×) 1
10
3.4(3y – 200) = 30(3y – 200) + (y – x)
10.2y – 680 = 9y – 600 + y – x
0.2y = 80 – x
2y = 800 – 10x
2y + 10x = 800………………..(ii)
Solve equation (i) and (ii) simultaneously:
3.5y + 10x = 950 200 + 10x = 800
2y + 10x = 800 10x = 600
1.5y = 150
y = 100 x = 60
Thus, the missing frequencies are:
Class intervals frequencies
20-30 x=60
20-40 y – 100
40-50 200 -60-100 = 40

WEEK 6: CONTINUOUS ASSESSMENT TEST (CAT) 25%

 Taken at Exam Centre

WEEK 7:

TOPIC: Measures of Dispersion

SUBTOPICS
o Range
o Inter-quartile range and quartile deviation
o Mean deviation and average deviation
o Standard deviation
o Variance

The degree to which numerical data vary (spread) about an average value is
called variation or dispersion of data. A small value for a measure of
dispersion indicates that the data are clustered closely, say, around the
arithmetic mean. The mean is therefore considered representative of the
data. That is, the mean is a reliable average. On the other hand, large
measure of dispersion indicates that the mean is not reliable; it is not
representative of the data.

Page 38
Various measures of dispersion or variation are available, the most common
being the range, mean deviation, quartile range, semi-interquartile range and
the standard deviation.

Significance of Measuring Variation

- To determine the reliability of an average.
- To serve as a basis for the control of the variability.
- To compare two or more series with regard to their variability.
- To facilitate the use of other statistical measures.

Properties of a good Measure of Variation

- It should be simple to understand.
- It should be easy to compute.
- It should be rigidly defined.
- It should be based on each and every item of the distribution.
- It should be amenable to further algebraic treatment.
- It should have a sampling stability.
- It should not be unduly affected by extreme items.

Range
Is defined as the difference between the value of smallest item and the value
of the largest item included in the distribution. Symbolically:
Range = L – S

Where: L = largest
S= smallest
The relative measure corresponding to range is called the coefficient of range
and is obtained by this formula:
L−S
Coefficient of range = L+S

Illustration
The following are the prices of shares of AB Company from Monday to
Saturday:
Day Price (£)
Monday 200
Tuesday 210
Wednesday 208
Thursday 160
Friday 220
Saturday 250
Calculate range and its coefficient.

Solution
Range = L – S
L = 250
S = 160

Page 39
 R = 250-160
= £90
L−S 250−160
Coefficient of range = L+S = 250+160 = 0.22

Computing Range in a continuous Series

Illustration
Calculate range and its coefficient from the following data:
Marks No. of Students
10-20 8
20-30 10
30-40 12
40-50 8
50-60 4
60−10
Coefficient of range = 60+10
50
= 70
= 0.714

Continuous Series
There are two methods of determining the range from data grouped into a
frequency distribution.
 The 1st method is to find the difference between the upper limit of the
highest class and the lower limit of the lowest class.
 The second method is to subtract the midpoint of the lowest wage
class from the midpoint of the highest wage class.

Merits of the Range

- Is simplest to understand and the easiest to compute.
- It takes minimal time to compute.

Limitations of Range
- Is not based on each and every item of the distribution.
- It is subject to fluctuations of considerable magnitude from sample to
sample.
- Range cannot tell us anything about the character of the distributions
within the two extreme observations.

Interquartile Range
It represents the difference between the 3 rd quartile and the 1st quartile
symbolically.
Inter-quartile Range = Q3 – Q1
Very often the inter-quartile range is reduced to the form of semi-inter-
quartile range or quartile deviation by dividing it by 2.

Page 40
 Q3 −Q1
Quartile Deviation or Q.D = 2
Quartile deviation gives the average amount by which the two quartiles differ
from the median. in asymmetrical distribution the two quartiles (Q1 and Q3)
are equidistant from the median i.e. Med – Q 1 = Q3 – Med, and as such the
difference can be taken as a measure of dispersion. The median ± Q.D
covers exactly 50% of the observations.

Coefficient of Quartile Deviation

Q 3 −Q 1
Coefficient of Q.D = Q 3 +Q 1
It can be used to compare the degree of variation in different distributions.

Illustration
Compute quartile deviation and its coefficient from the following data:
Wages in (£) No. of wage Cf
earners
Less than 35 14 14
35-37 62 76
38-40 99 175
41-43 18 193
Over 43 7 200

Q3 −Q1
Q.D = 2
N 200
Q1 = size of the 4 th item = 4 = 50th item
Q1 lies in the class 35-37

( )
N
4
−cf
i
Q1 = L + f

Q1 lies in the 35-37th

L = 35
N
4 = 50
Cf = 14
f = 63
i=2

Q1 = 35 +
(50−14
62 )×2 = 36.16

Page 41
 3N
Q3 = size of 4 th item
3 x 200
= 4
= 150th item
Q3 lies in the class 38-40

Q3 =
L+ (
i 3N
f 4
−c )
L = 38
3N
4 = 150
Cf = 76
f = 99
i= 2

Standard Deviation
It measures the absolute dispersion (or variability of distribution). The greater
the amount of dispersion or variability, the greater the standard deviation
since the greater will be the magnitude of the deviations of the values from
their mean. A small standard deviation means a high degree of uniformity of
the observations as well as homogeneity of a series; a large standard
deviation means just the opposite.
Computation Procedure
1. for ungrouped data
a) Sample standard deviation

√ ∑ ( x i−x ) 2

S= n−1
Where: xi= are the individual observations

x = mean of the series

n = sample size
b) Population standard deviation

√Σ fx2 Σ fx
( )
2
δ= −
Σf Σf
Where: x = midpoint of each class
f= frequency of each class

Page 42
Example:
The data below shows the weight of 6 students in a college.
72 48 70 40 52 60
Calculate the standard deviation.

Xi ( i
x −x ) ( xi−x )2
72 15 225
48 9 81
70 13 169
40 -17 289
52 -5 25
60 __ 3 9____________
Σxi =342 Σ ( xi−x )2=798

√ ∑ ( x i−x ) 2

S= n−1

798
= 6−1√
=
798
5 √
=√ 159 .6
= 12.63

Illustration
The data below relates to wage distribution in a large textile industry:
Wage(000’s) No. of Midpoint fx fx2
ksh employees (f) (x)

0-10 5 5 25 125

10-20 12 15 180 2700

20-30 30 25 750 18750

30-40 45 35 1575 55125

40-50 50 45 2250 101250

50-60 37 55 2035 111925

Page 43
60-70 21 65 1365 88725

∑f=200 ∑fx = 8180 ∑fx2 = 378

600

√
Σ fx2 Σ fx
( )
2
δ= −
Σf Σf

√√ 378600 8075
( )
2

−
= 200 200
= 1827.375−1630.141
= √ 197 .234
= 14.84
Combined Standard Deviation
Combined standard deviation is denoted by:
δ 12 and is computed as follows:

δ 12=

Where:
√
N 1 δ 21 + N 2 δ 22 + N 1 d 21 + N 2 d 22
N 1+ N 2
δ 12 = Combined standard deviation
δ 1 = Standard deviation of first group
δ 2 = Standard deviation of 2nd group

d1 =
|x 1 −x12|

d2 =
|x 2 −x 12|

The above formula can be extended to find out the standard deviation of
three or more groups, thus;

δ 123=
√
N 1 δ 21 + N 2 δ 22 + N 3 d 23 + N 1 d 21 + N 3 d 23
N 1+ N 2+ N 3
|x −x12|;
Where: d1 = 1

d2
|x −x 123|
= 2 d3 =
|x 3−x 123|

Illustration
The data below shows some particulars of the distribution of weight of male
and female employees in an organization

Males Female
s

Number 100 50

Mean 60kg 45kg

Page 44
weight

Variance 9 4

a) Find the standard deviation of the combined data.

b) Which of the two distributions is more variable?

Solution:

a) Combined SD =
δ 12=

We compute the combined mean

√
N 1 δ 21 + N 2 δ 22 + N 1 d 21 + N 2 d 22
N 1+ N 2

N 1 x 1+ N 2 x 2
x 12=
N 1+ N 2
100(60 )+50(45 )
= 100+50
= 55
Thus: N1 = 100
N2 = 50
δ1 = 9
δ2= 4

d1 =
|x 1−x12|=60−55
d2
|x −x 12|=45−55
= 2

Thus:
δ 12=
150 √
100(9 )2 +50( 4 )+100 (5)2

= 150
= 7.57
√
900+200+ 2500+5000

b) To find which distribution is more variable we compute coefficient of

variation
δ
×100
CV = x
The more the CV, the more variable the distribution is.
δ
×100
Thus: C.V (boys) = x
3
×100
= 60
= 5.00

Page 45
 δ
×100
C.V (girls) = x
2
×100
= 45
= 4.44
Since CV is more for distribution of weight of boys, the distribution shows
greater variability.

Variance
The variance is a measure of how broadly distributed the random
variable tends to be. It’s defined in terms of the expected value:
Var (X) = ∑((X-E(x)) 2)
The variance is often denoted δ2, and its positive square root, δ, is
known as the standard deviation.

WEEK 8:

TOPIC: Measures of Dispersion

SUBTOPICS:
 Lorenz Curve
 Skewness
 Kurtosis

Lorenz Curve
This measure of dispersion is graphical. It is named after Dr. Max Lorenz. It is
generally used to show the extent of concentration of income and wealth.
The steps involved in plotting the Lorenz curve are:
(i) Convert a frequency distribution into a cumulative frequency table.
(ii) Calculate percentage for each item taking the total equal to 100.
(iii) Choose a suitable scale and plot the cumulative percentages of the
persons and income. Use the horizontal axis of X to depict
percentages of persons and the vertical axis of Y to depict
percentages of income.
(iv) Show the line of equal distribution, which will join 0 of X-axis with
100 of Y axis.
(v) The curve obtained in (3) above can now be compared with the
straight line of equal distribution obtained in (4) above. If the
Lorenz curve is close to the line of equal distribution, then it implies
that the dispersion is much less. If, on the contrary, the Lorenz
curve is farther away from the line of equal distribution, it implies
that the dispersion is considerable.
The Lorenz curve is a simple graphical device to show the disparities of
distribution in any phenomenon. It is, used in business and economics to
represent inequalities in income, wealth, production, savings, and so on.

Page 46
The above shows a Lorenz curve. The straight line BD is a line of equal
distribution, whereas BC shows inequality. Curve BC is a Lorenz curve.

In the Lorenz curves figure as above, as the first curve is nearer to the line of
equal distribution, it has more equitable distribution of income after taxes
than the second curve of income before taxes and benefits. Prior to taxation
and benefits, the curve shows greater inequality in the income of the
households. After taxation and benefits, the company’s data results into a
curve, this is closer to the line of equal distribution. In other words, as a result
of taxation and benefits, the inequality has reduced.

Skewness
Measures of skewness help us to distinguish between different types of
distributions. Two distributions may have the same mean and standard
deviation but may differ widely in their overall appearance. The term
'skewness' refers to lack of symmetry" i.e. when a distribution is not
symmetrical (or is asymmetrical) it is called a skewed distribution.

Page 47
Symmetrical Distribution: In symmetrical distribution, the values of mean,
median and mode coincide. The spread of the frequencies is the same on
both sides of the centre point of the curve as shown in the diagram below.

Asymmetrical Distribution: A distribution, which is not symmetrical, is

called a skewed distribution and such a distribution could either be negatively
or positively skewed as presented in diagram (a) and (c) below.

Positively Skewed Distribution: In the positively skewed distribution the

value of the mean is maximum and that of mode least-the median lies in
between the two.

Negatively Skewed Distribution: In a negatively skewed distribution the

value of mode is maximum and that of mean least-the median lies in
between the two. In the positively skewed distribution the frequencies are
spread out over a greater range of values on the high-value end of the curve
(the right-hand side) than they are on the low-value end. In the negatively
skewed distribution the position is reversed, i.e. the excess tail is on the left-
hand side.

In moderately symmetrical distributions the interval between the mean and

the median is approximately one-third of the interval between the mean and

Page 48
the mode. It is this relationship, which provides a means of measuring the
degree of skewness.

Tests of Skewness
In order to ascertain whether a distribution is skewed or not the following
tests may be applied. Skewness is present if:
(i) The values of mean, median and mode do not coincide.
(ii) When the data are plotted on a graph they do not give the
normal bell shaped form i.e. when cut along a vertical line
through the centre the two halves are not equal.
(iii) The sum of the positive deviations from the median is not equal
to the sum of the negative deviations.
(iv) Quartiles are not equidistant from the median.
(v) Frequencies are not equally distributed at points of equal
deviation from the mode.
On the contrary, when skewness is absent, i.e. in case of a symmetrical
distribution, the following conditions are satisfied:
(i) The values of mean, median and mode coincide.
(ii) Data when plotted on a graph give the normal bell-shaped form.
(iii) Sum of the positive deviations from the median is equal to the
sum of the negative deviations.
(iv) Quartiles are equidistant from the median.
(v) Frequencies are equally distributed at points of equal deviations
from the mode.

The formula for measuring skewness as given by Karl Pearson is as follows:

Skewness = Mean – Mode
Coefficient of skewness =(Mean–Mode)/Standard Deviation

In case the mode is indeterminate, the coefficient of skewness is:

Mean - (3 Median - 2 Mean)/ Standard Deviation
3(Mean – Median)/ Standard Deviation

The direction of skewness is determined by ascertaining whether the mean is

greater than the mode or less than the mode. If it is greater than the mode,
then skewness is positive. But when the mean is less than the mode, it is
negative. The difference between the mean and mode indicates the extent of
departure from symmetry. It is measured in standard deviation units, which
provide a measure independent of the unit of measurement.

The value of coefficient of skewness is zero, when the distribution is

symmetrical. Normally, this coefficient of skewness lies between +1. If the
mean is greater than the mode, then the coefficient of skewness will be
positive, otherwise negative.

Kurtosis
Kurtosis is another measure of the shape of a frequency curve. It is a Greek
word, which means bulginess. While skewness signifies the extent of
asymmetry, kurtosis measures the degree of peakedness of a frequency

Page 49
distribution. Karl Pearson classified curves into three types on the basis of the
shape of their peaks. These are mesokurtic, leptokurtic and platykurtic.

From the diagram below, mesokurtic curve (normal curve) is neither too
much flattened nor too much peaked. In fact, this is the frequency curve of a
normal distribution. Leptokurtic curve is a more peaked than the normal
curve. In contrast, platykurtic is a relatively flat curve.

The coefficient of kurtosis as given by Karl Pearson is β 2= (μ4/μ2)2. In case of a

normal distribution, that is, mesokurtic curve, the value of β 2=3. If β2 turn out
to be>3, the curve is called a leptokurtic curve and is more peaked than the
normal curve. Again, when β 2<3, the curve is called a platykurtic curve and is
less peaked than the normal curve. The measure of kurtosis is very helpful in
the selection of an appropriate average. For example, for normal distribution,
mean is most appropriate; for a leptokurtic distribution, median is most
appropriate; and for platykurtic distribution, the quartile range is most
appropriate.

WEEK 9:

TOPIC: Correlation Analysis

SUBTOPICS:
 Significance of Studying Correlation Analysis
 Scatter Diagram

Correlation analysis deals with the association between two or more

variables. Correlation analysis attempts to determine the degree of
relationship between variables. The problem of analysing the relation
between different series should be broken into three steps:

Page 50
a) Determining whether a relationship exists and if it does, measuring it.
b) Testing whether it is significant
c) Establishing the cause and effect relation if any.

Why study Correlation?

1. Most of the variables show some kind of relationship e.g. there is
relationship between price and supply, income and expenditure e.t.c.
By help of correlation analysis, we can measure in one figure the
degree of relationship existing between the variables.
2. Once we know that two variables are closely related, we can estimate
the value of one variable given the value of another. This can be done
using regression analysis.
3. Correlation analysis contributes to the understanding of economic
behavior, aids in locating the critically important variables on which
others depend, may reveal to the economist the connection by which
disturbances spread and suggest to him the paths through which
stabilizing forces may become effective.
4. In business correlation analysis enables the executive to estimate
costs, sales, prices and other variables on the basis of some other
series with which these costs, sales or prices may be functionally
related.

Types of Correlation
 Positive or negative
 Simple, partial and multiple
 Linear and non-linear

Positive and Negative Correlation

If both variables are varying in the same direction i.e. if as one variable is
increasing the other on an average is also increasing, or if as one variable is
decreasing, the other on average is also decreasing, correlation is said to be
positive.
If the variables are varying in opposite directions, i.e. as one variable is
increasing, the other is decreasing or vice versa, correlation is said to be
negative.

Simple, Partial and Multiple Correlations

The distinction between simple, partial and multiple correlations are based
upon the number of variables studied. When only two variables are studied, it
is a problem of simple correlation. When three or more variables are studied,
it is a problem of either multiple or partial correlation.
In multiple correlations three or more variables are studied simultaneously
e.g. when we study the relationship between the yield of rice per acre and
both the amount of rainfall and the amount of fertilizers used.
On partial correlation we recognize more than two variables, but consider only
two variables to be influencing each other, the effect of other influencing
variables being kept constant.

Linear and non-linear (curvilinear) correlation

Page 51
The distinction is based upon the constancy of the ratio of change between
variables. If the amount of change in one variable tends to bear constant ratio
to the amount of change in the other variable then the correlation is said to be
linear.
If the amount of change in one variable does not bear a constant ratio to the
amount of change in the other variable, correlation is called curvilinear.

Approaches of studying Correlation

 Scatter diagram method
 Karl Pearson’s coefficient of correlation
 Spearman’s rank correlation coefficient
 Regression analysis (method of least square)

Scatter Diagram Method

The simplest device for ascertaining whether two variables are related is to
prepare a dot chart called scatter diagram. When this method is used, the
given data are plotted on a graph paper in the form of dots, i.e. for each pair
of x and y values, we put a dot and thus obtain as many points as the
number of observations. By looking at the scatter of the various points we
can form an idea as to whether the variables are related or not. The greater
the scatter of the plotted points on the chart, the lesser is the relationship
between the two variables. The more closely the points come to a straight
falling from the lower left hand corner to the upper right-hand corner,
correlation is said to be perfectly positive. Depending on the nature of
correlation, we can have the following diagrams:

Page 52
Merits of the Method:
- It is a simple and non-mathematical method of studying correlation
between the variables. As such, it can be easily understood and a
rough idea can very quickly be formed as to whether or not the
variables are related.
- It is not influenced by the size of extreme items whereas most of the
mathematical methods of finding correlation are influenced by extreme
items.
- Making a scatter diagram usually is the first step in investigating the
relationship between two variables.

Demerits
- By applying this method, we can get an idea about the direction of
correlation and also whether it is low or high. But we cannot establish
the exact degree of correlation between the variables as it is possible
by applying the mathematical methods.

WEEK 10:

TOPIC: Correlation Analysis

SUBTOPICS:
 Karl Pearson’s Correlation Analysis
 Rank Correlation Analysis

Karl Pearson’s Product Moment Coefficient of Correlation

Page 53
Is the most widely used method in practice. It is denoted by the symbol r. It
is one of the very few symbols that are used universally for describing the
degree of correlation between two series. The formula is:

Where n = number of pairs of observations

The value of r obtained lies between ±1. When r = +1, it means there is
perfect positive correlation between the variables. When r = -1, it means
there is perfect negative correlation between the variables. When r = 0 it
means there is no relationship between the two variables. However, in
practice such values of r as +1, -1 and 0 are rare. We normally get values
which lie between +1 and -1 such as +0.8, -0.26, e.t.c. The coefficient of
correlation describes not only the magnitude of correlation but also its
direction. Thus, +0.8 means that correlation is positive and -0.26 means that
correlation is negative.

r=
∑ xy
r can also be computed this formula: √ ∑ x2 ∑ y2
Where x = X −X and y = Y −Y

Illustration
The following table gives indices of industrial production of registered
unemployed (in hundred thousand). Calculate Pearson’s coefficient of
correlation.
Year: 2010 2011 2012 2013 2014 2015 2016 2017

Index of 100 102 104 107 105 112 103 99

Production.

No. 15 12 13 11 12 12 19 26
Unemployed
:

Method 1
Production Unemployed
x x2 y y2 xy
100 10000 15 225 1500
102 10404 12 144 1224
104 10816 13 169 1352
107 11449 11 121 1177
105 11025 12 144 1260
112 12544 12 144 1344
103 10609 19 361 1957
99 9801 26 676 2574

Page 54
Σx= Σx2= Σy= Σy2= Σxy=
832 86648 120 1984 12388
nΣ xy−ΣxΣy

r= √[ nΣx −( Σx ) ] [ nΣy −( Σy ) ]
2 2 2 2

8×12388−832×20
= √[8×86648−8322 ][8×1984−1202 ]
−736
= √ 960×1472
= -0.619

This would imply a high degree of negative correlation between index of

production and unemployment level.

Method 2

Year X x−x Y y− y y− y
x x2 xy
y y2
2000 100 -4 16 15 0 0 0
2001 102 -2 4 12 -3 9 +6
2002 104 0 0 13 -2 4 0
2003 107 +3 9 11 -4 16 -12
2004 105 +1 1 12 -3 9 -3
2005 112 +8 64 12 -3 9 -24
2006 103 -1 1 19 +4 16 -4
2007 99 -5 25 26 +11 121 -55
ΣX= Σx= Σx2= ΣY = Σy= Σy2=184 Σxy =
832 0 120 120 0 -92

r=
∑ xy
√ ∑ x2 ∑ y2 −92
= √ 120×184
= -0.619 note, the two methods yield the same answer.

Page 55
Assumptions of the Pearson Coefficient
Karl Pearson’s coefficient of correlation is based on the following
assumptions:
 There is linear relationship between the variables, i.e., when the two
variables are plotted on a scatter diagram a straight line will be formed
by the points so plotted.
 The two variables under study are affected by a large number of
independent causes so as to form a normal distribution. Variables like
height, weight, price, demand, and supply e.t.c are affected by such
forces that a normal distribution is formed.
 There is a cause and effect relationship between the forces affecting
the distribution of the items in the two series. If such a relationship is
not formed between the variables, i.e., if the variables are
independent, there cannot be any correlation e.g. there is no
relationship between income and height because the forces that affect
these variables are not common.

Merits of the Pearson Coefficient

- The coefficient summarizes in one figure not only the degree of
correlation but also the direction, i.e. whether correlation is positive or
negative.

Limitations
- The correlation coefficient always assumes linear relationship
regardless of the fact whether that assumption is correct or not.
- Great care must be exercised in interpreting the value of this
coefficient as very often the coefficient is misinterpreted.
- The value of the coefficient is unduly affected by the extreme items.
- As compared with other methods this method takes more time to
compute the value of correlation coefficient.

Coefficient of Correlation and Probable Error

The probable error of the coefficient of correlation helps in interpreting its
value. With the help of probable error, it is possible to determine the
reliability of the value of coefficient in so far as it depends on the conditions
of random sampling. The probable error of the coefficient of correlation is
obtained as follows:
1−r 2
P.E = 0.6745 √ N
Where r is the coefficient of the correlation and N the number of pairs of
observations.
- If the value of r is less than the probable error there is no evidence of
correlation, i.e. the value of r is not at all significant.
- If the value of r is more than six times the probable error, the
coefficient of correlation is practically certain, i.e. the value of r is
significant.

Page 56
 - By adding and subtracting the value of probable error from the
coefficient of correlation we get respectively the upper and lower limits
within which coefficient of correlation in the population can be
expected to lie symbolically.
ρ=r±P . E
Where ρ (rho) denotes correlation in the population.

Illustration
Let us compute probable error, assuming a coefficient of correlation of 0.80
and a sample of 16 pairs of items. We will have:
1−(0 . 8 )2
P.E = 0.6745 √ 16
= 0.06
The limits of the correlation in the population would be r ± P.E
i.e. 0.8 ± 0.06 or [0.74, 0.86]

Conditions for the use of Probable Error

The measure of probable error can be properly used when following three
conditions exist:
1. The data must approximate a normal frequency curve.
2. The statistical measure for which the P.E. is computed must have been
calculated from a sample.
3. The sample must have been selected in an unbiased manner and the
individual items must be independent.

Illustration
If r=0.8 and N=36, find out the probable error of the coefficient of correlation
and determine the limits for population r.

Solution
1−r 2
P.E.r = 0.6745 √ N
1−0
= 0.6745 √ 36
= 0.04047
Limits of population correlation = 0.8± 0.04 = [0.76, 0.84]

Coefficient of Determination
The coefficient of determination equals r 2 and is a convenient and useful way
of interpreting the value of coefficient of correlation between two variables.
If r = 0.9 then r2 = 0.81 and this would mean that 81% of the variation in the
dependent variable has been explained by the independent variable. The
maximum value of r2 is unity because it is possible to explain all of the

Page 57
variation in dependent variable, but it is not possible to explain more than all
of it.
The coefficient of determination (r 2) is defined as the ratio of the explained
variance to the total variance.
totalvariance¿
explainedvariance¿ ¿
Coefficient of determination = ¿
The ratio of explained variance to total variance is frequently called
coefficient of non-determination. The coefficient to non-determination is
denoted by K2 and its square root is called the coefficient of alienation, or K.
The K and K2 values may also be used as the measure of the degree or
relationship between two variables. For example, the higher the unexplained
variance with respect to total variance, the higher will be the value of K 2 and
the value of K.

The coefficient of non-determination is one minus the coefficient of non-

determination. Symbolically,

K2 = 1 – r2
totalvariance¿
explainedvariance¿ ¿
=1- ¿
K= √ 1−
ExplainedVariance
TotalVariance

However, r2 and 4 are more convenient in interpreting the results of

correlation analysis. It is much easier to understand the meaning of r2 than 4
and, therefore, the coefficient of determination should be preferred in
presenting the results of correlation analysis.

Spearmans Rank Correlation Coefficient

It was proposed by the British psychologist Charles Edward Spearman in
1904. This measure is especially useful when quantitative measures for
certain factors (such as the evaluation of leadership ability or judgment of
female beauty) cannot be fixed but the individuals in the group can be
arranged in order there by obtaining for each individual a number indicating
his/her rank in the group.
Spearman’s Rank Correlation coefficient is defined as:

Where:
n= number of paired observations
=difference between ranks
In case of tied ranks, the formula is modified as follows:

Page 58
Where t= number of tied ranks

Illustration
Calculate Spearman’s coefficient of correlation between marks assigned to
ten students by judges x and y in a certain competitive test as shown below:
S. No. 1 2 3 4 5 6 7 8 9 10
Marks by x: 52 53 42 60 45 41 37 38 25 27
Marks by y: 65 68 43 38 77 48 35 30 25 50

Solution
We assign ranks and then calculate rank correlation coefficient.
Marks by Rx Marks by Ry Rx – Ry
Judge x Judge y d2
52 8 65 8 0
53 9 68 9 0
42 6 43 5 1
60 10 38 4 36
45 7 77 10 9
41 5 48 6 1
37 3 35 3 0
38 4 30 2 4
25 1 25 7 25
27 2 50 7 25
∑d2 =
76
6 Σd 2
1−
R= n(n 2−1 )
6×76
2
= 1 - 10(10 −1 )
= +0.539
There is a high positive correlation between the marks awarded by the two
judges.

Page 59
Merits of Rank Method
 The method is simpler to understand and easier to apply compared to
the Karl Pearson’s method. The answers obtained by this method and
the Karl Pearson’s method will be the same provided no value is
repeated, i.e. all the items are different.
 Where the data are of a qualitative nature like honesty, efficiency,
intelligence, e.t.c, this method can be used with great advantage. E.g.
the workers of two factories can be ranked in order of efficiency and
the degree of correlation established by applying this method.
 This is the only method that can be used where we are given the ranks
and not the actual data.
 Even where actual data are given, rank method can be applied for
ascertaining correlation.

Limitations:
 The method cannot be used for finding correlation in a grouped
frequency distribution.
 Where the number of items exceeds 30 the calculations become
tedious and require a lot of time. Therefore, this method should not be
applied where N exceeds 30 unless we are given the ranks and not the
actual values.

When to use Rank Correlation Coefficient

 The initial data are in the form of ranks.
 If N is fairly small (say, not more than 30).

WEEK 11:

TOPIC: Functions

SUBTOPICS:
 Concepts and definitions
 Types of Functions

Preliminary Concepts and Definitions

A function is a relationship in which values of a dependent variable are
determined by the values of one or more independent variable or variables.
We use the letter f to express a function. An example is: “Sales is a function
of quantity” is expressed as: Sales = f (quantity).

Types of mappings (Relationships)

Similarly, Savings = f (Income)
Demand f (Price, income, price of related goods, etc)

Before we define a function, there are certain preliminary ideas which we

require to understand.

Page 60
  Constant: is a quantity whose value remains unchanged throughout a
particular analysis e.g. a fixed cost such as salary or rent in a given
period.
 Variable:Is a quantity, which assumes or takes various values in a
particular analysis.

Examples
Suppose an item is sold for Shs.35 per unit and
Let S = Sales revenue
Q = Quantity sold
The price of Sh.35 is a constant but sales revenue (S) and quantity sold (Q)
are variables. However, these variables are of two types: independent
variable and dependent variable.

Independent and Dependent Variable

An independent variable is that which is capable of determining the quantity
or the value of some other variable, which is then termed the dependent
variable.
For the preceding illustration, quantity sold (Q) is the independent variable
whereas sales revenue (S) is the dependent variable.
Furthermore, since quantity sold “predicts” the sales revenue, which in turn
“responds” to sales quantity, Q is also called predictor variable and S the
response variable

Importance of Functions in Business

Functions are important in establishing relationships among business
variables. Knowledge of functions facilitates their control to enable the
realization of organizational objectives. This can include the making of
decisions in such areas as:
- Output required
- Prices to charge
- Level of advertising
- Optimal workforce size, etc.

Types of Functions
Some of the more common functions in business include the following:
 Constant functions
 Polynomial functions
 Multivariate functions
 Logarithmic functions
 Exponential functions
 Continuous versus Discrete functions
 Step functions
 Linear Functions

Constant Functions
A constant function has the same value for the dependent variable
irrespective of the values of the independent variable.

Page 61
Example is the function, y = 4x where x = independent variable and y =
dependent variable.
An example of a constant function in business is fixed costs (such as rent and
salaries) which do not change in the short run.
Explanation: “Short run” as used in business usually means a period within
one year.

Sketch of a Constant Function

Polynomial Functions
This is a function which takes the forms
y = a+ b1x +b2x2+ ………….bnxn
Where y = dependent variable
x = independent variable
a, b1, b2, bn = constants
n, the highest power of x, is known as the degree of the polynomial.
Examples of polnomials
1st degree polynomial (linear function) = → n=1
y a + bx
(l) y= l0 - 3x
(2) y = 20 + x
(3) y = 2x
2 degree polynoinial (Quadratic functions) → n = 2 and b2¿ 0.
nd

y = a + bx + b2x2
e.g.
(1) y = 25+2x—3x2
(2) y = 20 + 0.3x2
(3) y = x2
3 degree polynomial (Cubic functions) n = 3 and b 3 ¿ 0
rd

y = a + bx + b2x2 + b3x3
e.g
(1) y = 15-x+2x2+ l0x3
(2) y= x3
(3) y= x2-2x3

Page 62
In most of the business applications, we do not usually go beyond cubic
functions. As such, we will study only the first three types, i.e. linear,
quadratic and cubic functions.

Multivariate functions
These are functions with more than one independent variable.
 Functions with one independent variable are known as univariate
functions;
 Functions with two independent variables are known as bivariate
functions; while
 Functions with three or more independent variables are known as
multivariate functions.

Logarithmic Functions
These are functions where at least one of the terms is in logarithmic form.
This term could be in the independent or dependent variable or both, for
example:
(1) 3 log y = 7x
(2) y= 7 log2x
(3) 5 log 3y 6 log x

Example:
(1) Y = 3x - 4Z2 - Bivariate function (x and z are independent variables)
Y = 2x - 12z3 + 2w2 - Multivariate function (x, z and w are independent
variables).

Exponential Functions
These are functions where the predictor variable is at least part of an
exponent or power:
Examples
1) Y= l02x
2) Log y =15- 2x2 +12
3) 3 log y= 7x +2w- 3z2

Note: A function need not be classified in only one way. For example, number
3 above is, in addition to being an exponential function, also a multivariate
and a logarithmic function.

Continuous versus Discrete Functions

A continuous function is one in which the values of a dependent variable are
defined for all values of the independent variables. A discrete function, on the
other hand,, has its dependent variable defined only for particular values of
the independent variables i.e. gaps exist between values.

Example of continuous functions

Exercise:
Plot the graph of the function y = x2
X -5 -3 -1 1 3 5
Y 25 9 1 1 9 25

Page 63
From the graph, note that for y = x2 there is no gap.
Examples of discrete functions are:
1) y = x where x = number of children in a household
2) y x where x number of typographical errors in a textbook.
Step Functions:
These are functions which take a constant value (for a given range), and then
the constant changes e.g. fixed cost for production with increasing capacity.
e.g. The following function can be plotted graphically as follows:

Production Level(Units) Fixed Cost (Shs)

1—100 1000
101—1000 3000
1001—3000 4000
Beyond 3000 5000

Linear Functions

Page 64
The linear function is a polynomial of the form y = a + b x
Where
a = y intercept i.e. value of y when x 0
b = The slope or gradient i.e. change in y per unit change in
Sketches of the linear function

Properties of a Linear Function

These are the features or properties of the linear function. There are three
main ones:
i) It has only one solution (root) i.e. it can cross the x axis a maximum of
once.
Consider the general linear function:
y = a + bx
When y = 0
a + bx = 0
x= - a
b
ii) It has no turning (stationary/critical) point
iii) It is completely defined once when either:
- two points on the line are specified or;
- one point and the slope are given.

Page 65
WEEK 12:

TOPIC: Functions

SUBTOPICS:
 Application of Linear Functions in Business

Application of Linear Functions in Business

This section gives four different areas where functions are applied in
business. These are in the form of numerical examples and serve to
highlight how pervasive and important functions are in solving
business problems.

Computation of Commissions
A salesman’s daily income is composed of a fixed amount and a
variable component. The variable component is dependent on the
number of units sold. He finds that when he sells 10 units on a given
day, his income is Sh 600 whereas when sales double, his earnings
increase by only Sh. 100.
Determine:
i. Fixed daily earnings;
ii. Level of commission per unit sold and hence;
iii. His earnings when he sells 30 units.
iv. On a certain day the salesman is determined to earn Sh
3,500. Suppose that on the previous day he had guaranteed
orders of 20 units, how many more units must he sell in order
to achieve his target earnings?

Solution:
Let
E = total daily earnings
Q = total quantity of units sold in a day
E = a + bQ

To get the relationship between B and Q, we solve for the values of a

and b. When d= 10,E=600 and when Q=20,E=700.

Equation:
1. 600=a+0b
2. 700a+20b 600a+ 10(10)
600a+100 a600- l00
a=500
and substituting when a = 500; we have b = 10
Equation: E = 500 + IOQ

Page 66
Hence,
i) Fixed daily income, a = Sh.500
ii) Level of commission per unit sold b = Sh.10. This is the slope
iii) Salesman’s earnings when he sells 30 units:
Q = 30
E =500 + 10(30)
E = 500 + 300
E=Sh.800
iv) Expected or target earnings E = 3500 Let number of units
required be x
E = 3500
3500= 500+l0(20+x)
3000= 200+ l0x
3000—200= l0x
2800 =l0x
x = 2800/10 x = 280 units

Demand and Supply Functions and Market Equilibrium

Suppose a certain commodity has demand and supply functions (both
of which have been estimated to be linear) going through the following
points:
(i) When Price, P Sh 7500, Quantity, q 1000 units
P= Sh 4625, q= 750 units
(ii) When P Sh 2525, q 100 units
P= Sh 1525, q 200 units
You are required to:
(a) Obtain the linear equations that go through the points given in (i)
and (ii) above and clearly explain which the supply is and which the
demand function is.
(Assume this is a normal commodity).

(b)Explain what is meant by the term “market equilibrium”. Obtain the

market equilibrium for the above. Show your results on a graphical
sketch.
Solution:
(1) P=a+bq where P price Q=quantity
7500= a +1000b………………. (1)
4625a±750b ……………………(2)
Solving for (1) and (2), we have;
2875 = 250b
And b= 11.5
Substituting for a in (i), we have:
7500 a+ 1000(11.5)
7500= a+ 11500
a = -4000

Page 67
The General equation is
P=-4000+ 1l.5Q

Comment:
This is a supply function since it has a positive slope i.e. b 11.5
(2) Substituting:
p a + bQ
(1) 2525a+lOOb
2) 1525a+200b
Solving simultaneously, we get b -10
Substituting for a in (1) using the value of b, we have:
2525a+ 100 (-10)
We find a = 3525
Since b -10, we conclude that this is a demand function (the slope is
negative).
Equation P 3525- 1OQ

b) Market equilibrium is the point where demand equals supply for a

given commodity. It is indicated on the graph by the intersection of the
demand and supply curves.

Qd = 3525-10Q
Qs = -4000+11.5Q P

Pe= equilibrium price

Qe = equilibrium quantity

At market equilibrium: demand function = supply function

Qd = Ss
3525 – 10 Q=-4000 +11.5Q
Since prices are equal 3525 + 4000 = 1 1.5Q + 10Q
7525=21.5Q
Q = 7525
21.5

Page 68
Q=350 units

Price at equilibrium Pe = 3525 - 10Q

Pe = 3525 - 10(350)
Pe =3525- 3500
Pe=Sh.25

Fixed Assets Accounting, Straight Line Method of Depreciation

Tombe Transporters Ltd depreciates its fleet of trucks using a straight
line method. The current accounting year is coming to an end and
external auditors are examining the books of accounts. However, they
cannot get complete records concerning a truck which was acquired
3years ago. Its current book value is Sh 1,800,000 while its purchase
cost was Sh 4,200,000. This type of truck is usually disposed off after 5
years.

Required:
a) Determine the linear function v = a + bt which relates the book
value v and time t in years. Interpret a and b.
b) What is the book value at the end of the 2 year of the truck?
c) Determine the disposal value of the truck stating any
assumptions you may make.

Solution
Equation 1: 4,200,000 = a + b(0)
a Sh.4,200,000 …………………………purchase price

Substituting
1800000 = 420000 - 3b
1800000 - 420000 = 3b
2400000 = -3b
b =2400000
-3
b =-800000
b = Sh 800,000

Equation
V = 4200000 - 800000t
b = Sh. 800,000 = annual depreciation (the value is negative since
vehicle loses value each year)

Page 69
b) When t = 2
v = 4.2- 0.8(2)
v= 4.2- 1. 6
v = Sh.2.6 million

c ) When t = 5 v=4.2-0.8(5)
= 4.2 - 4
v = 0.2 v = Sh.200, 000

Assumption: The disposal value is the same as the book value.

Cost-Volume Profit (C-V-P) Analysis
This is also known as profit planning and breakeven analysis. Profit is a
function of prices, costs, volume and other factors.

Problem
How does management manipulate factors which determine profits in
order to achieve the objective of maximizing profits?

For linear C-V-P analysis, we make the following assumptions.

i) Linearity. We assume that the revenue, cost and hence profit
functions are linear with respect to the level of activity i.e. (quantity
produced and sold)
- For revenue function to be linear, price per unit must be
constant e.g. there should be no quantity discounts.
- For cost function to be linear:
- unit variable costs are constant e.g. there are no changes
in direct material or direct labour costs.
- Fixed costs remain constant.
ii) All costs can be classified as either fixed or variable. In particular,
there are no semi-variable costs.
iii) All units produced are sold i.e. inter-period inventory changes are
negligible.
iv) The only factor which influences revenues, cost and hence profits is
level of activity.

Page 70
v) There is no demand restriction i.e. there are no constraints as to the
quantity that can be sold.
vi) All factors which influence profits are known in advance with
certainty.

Equation approach
Let R = total revenue in monetary terms
P = unit price
V = unit variable costs
X = sales quantity (physical units)
f = fixed cost
C = total cost
∏ ¿¿= profit
V = total variable costs
a) Sales in physical units (x)
i. Profit equation
By definition:
=R-C
R = px
p - v = Contribution margin
C = vx + f
= px- (vx +f)
 = px – vx -f
= (p- v) x - f
 = cm*x - f
CM = Contribution margin per unit
2. Unit sales X for target profit, T (Sh)
From equation above,  = T
Hence T = CM * —f
Making x the subject, we have:
x = T +f
CM
3. Breakeven sales units, X b/e (b/e = breakeven)
At B.E.P, R = C so that ∏ ¿¿= 0
From equation 2 above T at B.E.P = 0
x(b/e) = f
CM

Graphical Representation

Page 71
Let actual sales (beyond breakeven sales) be Xa
Xa - Xb the extent by which actual sales exceeds breakeven sales is
known as Margin of Safety (M0S).

This is also the extent by which sales would have to fall before the firm
begins to make losses.

Example 1: Sales in Physical Units

A product has a selling price of Sh. 200 whereas unit variable cost is
Sh. 140. The annual fixed cost is Sh. 720,000. You are required to
determine the following:
a) breakeven sales in units;
b) profit to be made if 20000 units are sold;
c) Sales required for a profit of Sh. 2,000,000.

Solution:
B.E.P. is given by the formula:
b) E ( in units) = fixed cost CM = p-v 200—140 = Sh60
CM
= 720,000 = 12000 units
60
b) The profit function is:
∏ ¿¿= 60 x 20000- 720000 Sh.480,000.
c) If the target profit is Sh T, then units, x, required to make this profit
is given as:
f +T
x=
CM
x = 720000 + 2000000 = 2,720,000
60 60
x = 45333 units (to the nearest whole number).

Example 2 — Sales in Monetary Units

A firm sells a product whose data in two periods is as follows:
Period Sales (Sh) Variable cost Profit (Sh)

Page 72
 (Sh)
I 100000 60000 20000
II 150000 90000 40000

Assume the price, unit variable cost, and fixed costs are the same in
the two periods.
Required:
a) Determine the fixed cost;
b) Determine the breakeven sales revenue;
c) What is the profit when sales are Sb. 600000?
d) What is the sales required for a profit of Sh.1 10000?
e) Determine the profit if variable cost incurred is Sh. 300000

Solution:
a) When sales are in monetary terms, profit function is given as:
∏ ¿¿ =CMRX(R-f)
We substitute the two given points to get the following equations:
(1) 20000 =CMR X(100000-f)
(2) 40000 =CMR X (150000-f)
To eliminate f, we subtract thus: (2) -(1):
20000 = 50000 CMR

CMR=20000
50000
CMR = 0.4
Using equation (1):
20000=0.4 X 100000-f
20000 =0.4 X 100000-f
Or:
f=-20000+0.4 x 100000
f= Sh. 20000

b) B.E.P.= f
CMR

B.E.P. = 20000
0.4
= sh. 50000
c) ForRsh.600000
∏ ¿¿= 0.4x600000- 20000
= sh. 220000

d) For a target profit T, sales revenue required:

R= f + T
CMR

Page 73
 R= 20000+110000
0.4
R= 130000
0.4
R = Sh. 325000
e) To answer this part, we need to know the sales revenue which is
consistent with a variable cost of sh. 300,000. We use the
equation:
CMR + VCR = 1 and since CMR 0.6, it follows that VCR = 0.6
300000 = 0.6/R

= R = 300000
0.6
R=Sh 500000
Hence profit, ∏ ¿¿= 0.4 x 500000 - 20000
=Sh 180000

WEEK 13:

TOPIC: Regression Analysis

SUBTOPICS:
 Definitions of Regression Analysis
 Uses of Regression Analysis
 Regression Lines

Definitions of Regression Analysis

1. Regression is the measure of the average relationship between two or
more variables in terms of the original units of the data.
2. The term ‘regression analysis’ refers to the methods by which
estimates are made of the values of a variable from a knowledge of the
values of one or more variables and to the measurement of the
errors involved in the estimation process.
3. One of the most frequently used techniques in economics and business
research to find a relation between two or more variables that are
related casually is regression analysis.
4. Regression analysis attempts to establish the nature of relationship
between variables –that is to study the functional relationship
between the variables and thereby provide a mechanism for
prediction, or forecasting.

From the above definitions it is clear that regression analysis is a statistical

device with the help of which we are in a position to estimate (or predict) the
unknown values of one variable from the known values of another variable.
The variable which is used to predict the variable of interest is called the
independent variable or explanatory variable and the variable we are trying

Page 74
to predict is called the dependent variable or the ‘explained’ variable, or the
‘response’ variable.

The independent variable is denoted by x and the dependent variable by y.

The analysis used is called the simple (bivariate) regression analysis – simple
because there is only one predictor or independent variable, and linear
because of the assumed linear relationship between the dependent and the
independent variable.

The term ‘linear’ means that an equation of a straight line of the form y = a
+ bx, where a and b are constants, is used to describe the average
relationship that exists between the two variables.

Uses of Regression Analysis

Regression analysis is a branch of statistical theory that is widely used in
almost all the scientific disciplines. In economics it is the basic technique for
measuring or estimating the relationship among economic variables that
constitute the essence of economic theory and economic life. For example, if
we know that two variables, price (x) and demand (y) are closely related, we
can find out the most probable value of x for a given value of y, or the most
probable value of y for a given value of x. Thus the study of regression is of
considerable help to the economists and businessmen.
Regression analysis attempts to accomplish the following:
- Regression analysis provides estimates of values of the dependent
variable from values of the independent variable. The device used to
accomplish this estimation procedure is the regression line. The
regression line describes the average relationship existing between x
and y variables, i.e. it displays mean values of x for given values of y.
- Regression analysis obtained a measure of the error involved in using
the regression line as a basis for estimation. For this purpose the
standard effort of estimate is calculated. This is a measure of the
scatter or spread of the observed values of y around the corresponding
values estimated from the regression line. If the line fits, the data
closely, that is, if there is little scatter of observation around the
regression line, good estimates can be made of the y variable and vice
versa.
- With the help of regression coefficients we can calculate the
correlation coefficient. The square of correlation coefficient (r), called
coefficient of determination, measures the degree of association of
correlation that exists between the two variables. It assesses the
proportion of variance in the dependent variable that has been
accounted for by the regression equation. In general, the greater the
value of r2, the better is the fit and the more useful the regression
equations as a predictive device.

Regression Lines
If we have two variables x and y, then we shall have two regression lines and
consequently two regression equations.
i) The regression equation of y on x denoted by y = a + bx
Where:

Page 75
 y = dependent variable
x = independent variable
a and b are constants to be determined in the equation.
ii) The regression equation of x on y denoted by: x = a + by
Where:
x = dependent variable
y = independent variable
a and b are constants to be determined in the equation.

*Given the values of x and y, if we need to generate the equations we have

to determine the constants as follows:

a) In case of the regression equation of the form Y = a + bx

The constants a and b can be obtained as follows;
Σy Σx
a= −b
n n

We can also obtain the values of a and b by solving the following normal
equations simultaneously
Σy=Na+bΣx
Σ xy=aΣx+bΣx2
b) In case of the regression equation of the form X = a + by
we obtain a and b as follows

Again, we can solve the following normal equations simultaneously

Σy=Na+bΣy
Σ xy=aΣy+bΣy2
*When there is either perfect positive or perfect negative correlation between
the two variables (r = ± 1) the regression lines will coincide, i.e. we will have
only one line.
Example:
Consider the data on index of production and number unemployedon page
75. We want to obtain the two regression equations
i) Y = a + bx
x y x2 y2 xy
100 15 1000 225 1500
102 12 10404 144 1224
104 13 10816 169 1352
107 11 11449 121 1177
105 12 11025 144 1260
112 12 12544 144 1344
103 19 10609 361 1957

Page 76
 99 25 9801 676 2574
832 120 86648 1984 12388

nΣ xy −ΣxΣy Σy Σx
2 2 a= −b
b = nΣx −( Σx ) n n
8×12388−832×120 120 832
−0 .77
= 8×86648−8322 = 8 8
−736
= 960 = 15 + 79.23
= -0.766666 = 94.73
≈ -0.77 y = 94.73 – 0.77x
With this equation, we can estimate the value of y given any value of x. e.g.
Estimate y when x = 120.

Therefore: since
Y = 94.73 – 0.77x
Y = 94.73 – 0.77 x 150
Y = 2.33
≈ 2 people
ii) X = a + by
nΣ xy −ΣxΣy
2 2
b = nΣy −( Σy )
8×12388−832×120
= 8×1984−1202
−736
= 1472
= -0.5
Σx Σy
−b
a= n n
832 120
−−0 .5
= 8 8
= 104 + 7.5
= 111.5
Therefore x = 111.5 – 0.5y

Therefore, when y = 30
x= 111.5 – 0.5 x 30
x = 11.5 – 15
x = 96.5

WEEK 14:

TOPIC: Matrices

Page 77
SUBTOPICS:
 Definition
 Order of Matrix
 Types of Matrices

Definition
A matrix is any rectangular array of numbers, written in the form:

[A=¿ a1 a12. .aln¿][a21a2 . .a2n¿][: : :¿]¿¿

¿ or A = (aij)

Aijs stand for real numbers and are the elements of the matrix.
In particular, a is the element in the ith row and the jth column of the matrix
A.
e.g

A=¿ [1 2¿] ¿ ¿¿ B=¿ [ 1 2 3 ¿ ] ¿ ¿¿ C=¿ [ 3 4¿ ] ¿ ¿¿

¿ ¿ ¿ are all matrices

Note
(1) A matrix is denoted using capital letters
(2) Each number in the matrix is an element a11 of the matrix where i
represent the row position and j represents the column position.

Order of the Matrix

This is defined by the number of rows and columns in a matrix. Thus, if a
matrix has m rows and columns it is said to be an m x n (read ‘m by n’)
matrix or a matrix of order m x n.. The order of matrices is frequently
denoted by subscripts e.g. Am xn

For example, the matrix,

A=¿ [1 2 6 ¿ ] ¿ ¿¿
¿ is of order 2x3 since ith as rows and three columns.

Similarly the matrix,

6 6 3 1
B= 3 8 -2 7 is of order 3 x 4 since it has three rows and 4 columns.
1 4 5 0

Types of Matrices

Vectors
Row vector is a matrix with only one row. E.g. V 1=(l0 -3 4 6) is a 1x4 row
vector

Page 78
 3¿ [ 7¿ ] ¿ ¿
Column vector is a matrix with only one column e.g. V 2= ¿
is a 2x1 column vector.

Square Matrix
If the number of rows and the number of columns of a matrix are equal, the
matrix is said to be a square matrix. i.e. if in n, the matrix is a square matrix.

S1=¿ [2 −5 3¿ ][7 4 6¿ ]¿¿¿

(i) ¿ is a 3x3 square matrix
S2=¿ [ 3 7 ¿ ] ¿ ¿¿
¿ is a 2x2 square matrix
Sub matrix
This is a matrix which is formed from any matrix by striking out one (or more)
rows or columns from the original matrix.

=¿ [5 4 9 2¿ ][10 4 3 −1¿] ¿ ¿¿
Let the original matrix be P. If P ¿
Sub-matrices:

B1=¿ [ 4 9 2¿] [4 3 −1¿ ]¿ ¿¿

¿ st
1 column omitted

B2=¿ [5 ¿ ][10¿ ]¿ ¿¿
¿ 2 , 3 and 4 columns omitted
nd rd th

B3=¿ [ 4 3 −1¿ ] ¿ ¿¿
¿ 1st column, 1st row omitted

Diagonal Matrix
This a square matrix whose every element other than at least one element in
the leading diagonal equals zero.
Examples

Page 79
 D1=¿ [ 1 0¿ ] ¿ ¿¿
i. ¿

D2=¿ [0 0 0¿] [0 −7 0¿]¿ ¿¿

ii. ¿
D3=¿ [ 0 0 0¿] [0 0 0¿ ] ¿ ¿¿
iii. ¿
Scalar matrix
A scalar matrix is a diagonal matrix whose leading diagonal elements are
equal
Examples

S1=¿ [−5 0¿ ] ¿ ¿¿
¿

S2=¿ [20 0 0¿] [0 20 0¿ ]¿ ¿¿

¿
Identity Matrix
A square matrix for which all the leading diagonal elements are equal to 1
and all other elements are equal to 0 is called an identity matrix. e.g

I 2=¿ [1 0¿] ¿ ¿¿
¿ 2x2 identity matrix

I3=¿[1 0 0¿] [0 1 0¿]¿¿¿

¿ 3x3 identity matrix
In matrix algebra, the identity matrix behaves like in real number algebra

Transpose of a Matrix
Let A be a matrix. If the rows and columns of the matrix A are interchanged -
so that the 1st row becomes the 1St column, the 2nd row becomes the 2nd
column and so on, then the new matrix formed is known as the transpose of
A, denoted A’ or AT.

A=¿ [1 2 3¿ ] ¿ ¿¿
For example, if ¿ then the transpose of A is

Page 80
A =¿ [ 1 4¿] [2 5 ¿ ] ¿ ¿¿
T
¿
Symmetric Matrix
If a matrix its transpose is equal, then the matrix is said to be symmetric
(about its main/leading diagonal)

M 1=¿ [ 3 0 ¿ ] ¿ ¿¿
¿

M 2=1¿ [ 5 7 1¿ ] [ 7 3 −2 ¿ ] ¿ ¿
¿

WEEK 15:

TOPIC: Matrix Operations and Algebra

SUBTOPICS:
 Equality of Matrices
 Addition of Matrices
 Subtraction of Matrices
 Multiplication of Matrices
 Determinants and Inverse of Matrices

This consists of ways of combining matrices in order to obtain other matrices

and values of interest. This gives rise to matrix algebra.

Equality of Matrices
Two matrices A (aij) and B = (bij) are equal if and only if A and B are of the
same order and aij = bij for all i and j.

Page 81
Addition and Subtraction of Matrices
The matrices to be added or subtracted must be of the same order. The
operations are carried out by adding or subtracting corresponding elements.
Examples
[ 1 −3 1¿ ] ¿ ¿¿ [ 20 40 −2¿ ] ¿ ¿¿
Let p = ¿ and Q= ¿
Find (i) P+Q (ii) P-Q (iii) Q-P

[ 1 −3 1¿ ] ¿ ¿¿ [ 20 40 −2¿ ] ¿ ¿¿
(i) ¿ ¿ +
[ 4+20 −3+40 1−2 ¿ ] ¿ ¿¿
=¿

[ 24 3 −1 ¿ ] ¿ ¿¿
= ¿
[ 4 −3 1¿ ] ¿ ¿¿ [ 20 40 −2¿ ] ¿ ¿¿
(ii) P-Q =¿ ¿ -
[ 4−20 −3−40 1−2 ¿ ] ¿ ¿¿
=¿

[−16 −43 3 ¿] ¿ ¿¿
= ¿
[ 4 −3 1¿ ] ¿ ¿¿ [ 20 40 −1¿ ] ¿ ¿¿
(I) p = ¿ Q= ¿
[ 4 −3 1¿ ] ¿ ¿¿ [ 20 40 −1¿ ] ¿ ¿¿ [ 24 37 0 ¿] ¿ ¿¿
P+Q =¿ ¿
+ = ¿
[ 4 −3 1¿ ] ¿ ¿¿ [ 20 40 −1¿ ] ¿ ¿¿ [−16 −43 20 ¿ ] ¿ ¿¿
P-Q =¿ +¿ =¿

[ 20 40 −1¿ ] ¿ ¿¿ [ 4 −3 1¿ ] ¿ ¿¿ [−16 43 20 ¿ ] ¿ ¿¿
Q-P = ¿ - ¿ = ¿
Properties of matrix addition
i. Matrix addition is commutative i.e. A+BB+A
ii. Matrix addition is associative i.e. (A+B)+CA+(B+C)
iii. If  denotes the null or zero matrix of the same order as that of matrix
B, then

Page 82
 B+ =  + B= B

Multiplication of Matrices
Matrix multiplication can occur in one of two ways; by a scalar i.e. a real
number or by another matrix.

(i) Multiplication by a scalar

In this case we multiply every element in the matrix by a scalar
Example
Let R= [3 -2 5 ] and scalar, k = -0.7
-5 7 1
[ 3 −2 5 ¿] ¿ ¿¿ [−2.1 1.4 −3.5¿ ] ¿ ¿¿
Then k x R = 0.7 ¿ = ¿
Multiplication of two matrices
Two matrices can be multiplied if and only if the number of columns in one
matrix is equal to the number of rows in the other matrix. In particular, the
matrix product AB is 1efined if and only if the number of columns in A is the
same as the number of rows in B, in which case the matrices A and B are said
to be conformable for multiplication.
Further, if matrix A is of the order m x n and B is of the order n x p, then the
product AB will be of the order m x p i.e. it will have the same rows as A and
columns as B.

Illustration

[−3 5¿][−2−2¿]¿¿¿ [1 0¿][−3 −10 ¿]¿¿¿

If A= ¿ and B = ¿
[−3 5¿][−2−2¿]¿¿¿ [1 0¿][−3 −10 ¿]¿¿¿
AB = ¿ X ¿
[−3x1+5x3 −3x0+5x−10¿][−2x1±2x3 −2x10±2x−10¿]¿¿¿ [12 −50 ¿][−8 20 ¿]¿¿¿
= ¿ = ¿
Properties of Matrix multiplication
(i) Matrix multiplication is not necessarily commutative AB ≠BA. In fact BA
may not be defined. Thus the sequence of matrix multiplication must
always be preserved.
ii) Matrix multiplication is associative i.e. AB(C) = (AB)C
iii) Matrix multiplication is distributive i.e. A(B + C) = AB + AC
(iv) A = A=

Page 83
(v) If I is an identity matrix and matrix multiplication is defined, then
AI=IA = A

Determinants
To each square matrix A, there corresponds a single value known as its
determinant determined by A

Determinant of a 2x2 matrix

a 11 a12
Consider the general matrix A =
a 21 a22

Determinant of the matrix A is given as the product of the leading diagonal

elements minus the product of the secondary diagonal elements i.e.
Det A = a11 a22 – a21 a12

Note:
If the determinant of a matrix is equal to 0, it is said to be a singular matrix.
Otherwise it is non-singular.

Determinants of 3x3 matrices

Before we can discuss the determinant of a 3x3 matrix, we need to introduce
a concept known as a cofactor.

[a11 a12 a13 ¿][a21 a22 a23¿]¿¿¿

Consider the general 3x3 matrix A= ¿
Corresponding to each element aij of matrix A, there is a cofactor, C ij. A 3x3
matrix has nine elements, so there are nine cofactors to be computed. The
cofactor Cij,1 is defined to be the determinant of the 2x2 matrix obtained by
deleting row i and column j of the matrix A, multiplied by ‘+1’ or ‘-1’
according to the following pattern.

[+ − +¿][− + −¿]¿¿¿
¿
For example, suppose we wish to calculate C 23, which is the cofactor of
element a of the matrix A. The element a 23 lies in row 2 and column 3.
Consequently we delete 2nd row and 3rc column as indicated below:

[a11 a12 a13 ¿][a21 a22 a23¿]¿¿¿

¿
[ a11 a12 ¿] ¿ ¿¿
The 2x2 matrix whose elements have not been deleted is ¿
We obtain the determinant of this matrix and multiply by -1 since it is located
at a position with a negative sign in the pattern above.

Page 84
Thus the cofactor of a23 is: C23 = - 1(a11a32 - a31a12).
The same procedure is repeated for all the elements of the matrix.

Matrix inversion
Obtaining the inverse of a matrix is referred to as matrix inversion. Inverse by
adjoint of a matrix.
Step 1
From the matrix A, form a new matrix by replacing every element a by its
cofactor, which we may call C.
Step 2
Transpose this matrix (C) - writing rows as columns and columns as rows.
This new matrix is called the adjoint of A.
Step 3

Calculate the value of the determinant of A i.e. |A|

Step 4
The inverse is given by
1 T
A−1 = C
|A|

Example
Inverse of a 3 x 3 matrix
Let matrix be A.
Steps
• Obtain the determinant.
• Determine the matrix of minor elements, M (minor matrix)
• From M, above, obtain the matrix of Cofactor elements, C (cofactor
matrix)
• Transpose the cofactor matrix C to obtain C transpose, C T also called
the adjoint /adjunct matrix of A.
- The inverse of A will be:
1
A−1 =
|A|xC T
Examples:

[3 1 2¿] [0 5 1¿]¿¿¿
I) N1 = ¿ Determinant of N1 /N1/ = -51
- Minor element matrix M = ignore the row and column of

Page 85
e.g. element 3 i.e.

1 1 2
= -3 2 4 and remain with = -5 -2

[−3 -6 −24¿][2 -11 −10¿]¿¿¿

M= ¿
[+ − +¿][− + −¿]¿¿¿
Cofactor matrix C (multiply M by ¿
[−3 -6 −24¿][2 -11 −10¿]¿¿¿
C=M= M= ¿
[−3 -6 −9¿][6 -11 1¿] ¿¿¿
- Adjoint matrix CT ¿
1
T
Inverse matrix Ni -1
= ¿ NI / xC

−1
= ¿ [−3 -3 −9¿ ] [ 6 -11 1¿] ¿ ¿
51 ¿
7 -3 6
4 2 8
(ii) N3 = -1 0 2 determinant = N3/ = 88

[4 16 2¿][−6 20 3¿] ¿¿¿

a) Minor element matrix M = ¿
[4 6 −36¿][-16 20 −32¿]¿¿¿
b) Adjust (Transpose) matrix CT = ¿
Page 86
 1
T
c) Inverse matrix N3-1 = ¿ N 3 / xC

−1 [ 4 6 −36¿] [-16 20 -3 2¿ ] ¿
= ¿ ¿
88 ¿

Page 87