Indore Institute of Law

Unit III
Measures of Central Tendency
Prof. Ambarish Bapat
 Introduction
One of the major objectives of the statistical
analysis is to get the single value that describes
the characteristics of the entire mass of unwieldy
data. Such a value is called the central value or
an average or the expected value of the variable.
The word average is commonly used in day to
day conversation.
Definitions of Average –
Clark – “Average is an attempt to find one single
figure to describe whole of figures”.

A. E. Waugh – “An average is a single value

selected from a group of values to represent
them in some way – a value which is supposed to
stand for whole group, of which it is a part, as
typical of all the values in the group”.

Leabo – “The average is sometimes described as

a number which is typical of the whole group”.
Objectives of Averaging
a. To get a single value that describes the
characteristics of the entire group.
b. To facilitate comparison.
Requisites of a Good Average –
a. Easy to Understand – As is aims to simplify
complexity it is desirable for the average to be such
that can be easily understood.
b. Simple to Compute – An average should not only be
easy to understand but also simple to compute so
that it be used widely.
c. Based on all the Items – The average should depend
upon each and every item of the series so that if any
of the items is dropped the average itself is altered.
d. Not be unduly Affected by Extreme Observations –
Although each and every item should influence the
value of the average, none of the items should
influence it unduly.
e. Rigidly Defined – An average should be properly
defined so that it has one and only one
interpretation. It should preferably be defined by
algebraic formula so that if different people
compute the average from the same figures they get
the same answer.
f. Capable of Further Algebraic Treatment – It could
be used for further statistical computation so that
its utility is enhanced.
g. Sampling Stability – It means that it should be
compatible with different samples i.e. it should it
generate approximate results.
Types of Averages –
Following are the important types of averages –
1. Arithmetic mean – i. simple and ii. Weighted.
2. Median
3. Mode
4. Geometric Mean
5. Harmonic Mean

There are less important averages like moving average,

progressive average, etc. These averages have a very
limited field of application and hence are not suitable.
 Arithmetic Mean – It is the most popular and
widely used measure of representing the entire
data by one value. It is termed as ‘average’ in
general terms. Its value is obtained by adding
together all the items and by dividing this total
by the number of items.
It can be –
i. Simple Arithmetic Mean
ii. Weighted Arithmetic Mean
Calculating Simple Arithmetic Mean – Individual
Observations –
Direct Method - It is calculated by adding all the
items and dividing by total number items. It is
expressed as –
X = X1+X2+X3+……Xn / N
= ∑X
N
∑X = sum of all the values of the variable.
N = Number of observations.
X = Arithmetic Mean.
 Short Cut Method: The arithmetic mean can be
calculated using an assumed mean. After taking
deviations using assumed mean, following formula
is used to calculate mean –
X = A + ∑d
N
Here
A is the assumed mean
d is the deviation of items taken from assumed
mean
N is the total number of terms
Steps involved –
1. Take an assumed mean.
2. Take the deviations of items from the assumed
mean and denote the deviations by d.
3. Add the deviations to obtain sum, i.e. ∑d.
4. Apply the formula : X = A + ∑d/N.
Calculation of Arithmetic Mean – Discrete Series
Direct Method –
Formula –
X = ∑fX / N
f = Frequency
X = The variable in the question
N = Total number of observations = ∑f
Steps –
1. Multiply the frequency of each row with the
variable and obtain total ∑fX.
2. Divide the total obtained by N i.e. ∑f (total
frequency).
Short Cut Method –
Formula – X = A+∑fd / N
A is assumed mean
d is (X – A)
N is total number of observations i.e. frequency.
Steps –
1. Take an assumed mean i.e. A.
2. Take the deviations (X – A) and denote it by d.
3. Multiply it by respective frequency and obtain
∑fd.
4. Divide the total obtained by frequency (∑f) and
add / subtract to A (assumed mean).
 Calculation of Arithmetic Mean – Continuous
Series -
Direct Method –
X = ∑fm / N
m is midpoint of various classes
N is ∑f i.e. total of frequency.
Steps –
1. Obtain the midpoint of each class, denote by m.
2. Multiply these mid points by respective
frequency of each class and obtain ∑fm.
3. Divide the total obtained by the sum of the
frequency i.e. N.
Short – cut Method –
Formula – X = A + ∑fd / N
A is assumed mean
∑fd is summation of product of deviations and
mid-point.
N total of frequency (∑f)
Steps –
1. Take an assumed mean.
2. From the midpoint of each class deduct assumed
mean.
3. Multiply the deviations by respective frequency
and obtain ∑fd. Divide it by ∑f.
Step Deviation Method –
Formula – X = A + ∑fd x i
N
A is assumed mean
d is deviations
N total number of observations (∑f)
i is difference of class interval.
Steps –
1. Take an assumed mean.
2. From the midpoint of each class deduct assumed mean.
3. Multiply the deviations by respective frequency and obtain
∑fd. Divide it by N & multiply by i.
4. Add/subtract to assumed mean.
Merits – It has following advantages –
1. It is simple to understand and easiest to
calculate.
2. It is affected by the value of every item in the
series.
3. It is defined by a rigid mathematical formula, it
is useful for further algebraic computations
better than median or mode.
4. It is more reliable as it does not vary too much
when repeated samples are taken from the
same population.
5. It balances the values at the extremes.
Limitations – It has following limitations –
1. Incase of open-end classes the value of mean
cannot be calculated without making
assumptions regarding the size of the class
interval of the open-end classes.
2. It is not always the best measure of central
tendency.
Weighted Arithmetic Mean –
One of the shortcoming of the Arithmetic mean
is that it gives equal importance to all the items.
But there are cases where the relative
importance of the different items is not the
same. In such cases weighted arithmetic mean is
computed. It is calculated as under –
Xw = ∑WX
∑W
When no formal questionnaire is used,
interviewers adapt their questioning to each
interview as it progresses. When a prescribed
sequence of questions is followed, it is referred
to as structured study. Incase no prescribed
sequence of questions exists, the study is non-
structured.
When questionnaires are constructed so that the
objective is clear to the respondents they are
non-disguised. When the objective not clear the
questionnaire is disguised one.
 On the basis of these two criterion four types of studies can
be distinguished –
 Non-disguised structured.
 Non-disguised non-structured.
 Disguised structured.
 Disguised non-structured.

It has following advantaged –

1. It can be used where the field of investigation is very vast
and the informants are spread over a wide geographical area.
2. Relatively cheap and expeditious provided informant
responds on time.
3. This method is superior where questions of personal
nature or reactions of the respondents are needed.
Limitations –
1. It can be adopted where the informants are
literate people so that they can understand
written questions and send the answers in
writing.
2. It involves some uncertainty about the
response. Co-operation on the part of informants
may be difficult to presume.
3. The information supplied by the informants
may not be correct and it may be difficult to
verify the accuracy.
 Drafting the questionnaire –
While designing a questionnaire it is of utmost
importance to follow certain rules –
Covering letter – The person conducting survey
must introduce himself and state the objective of
the survey. It is desirable to have a short letter
enclosed stating the same.
Number of questions should be small.
Questions should be arranged logically.
Questions should be short and simple to
understand.
Ambiguous questions ought to be avoided.
Personal questions should be avoided.
Instructions to the informants.
Questions should be capable of objective answer.
Yes or No question.
Specific information questions and open-end
questions.
Questionnaire should look attractive.
Questions requiring mathematical calculations
should be avoided.
Pre-testing the questionnaire.
Cross-checks.
Method of tabulation.
 Classification of Data
Meaning – After collecting and editing data, the
first step towards further processing the same is
classification. It is the grouping of related facts
into classes. Facts in one class differ from those
of classification. Sorting facts on the basis of
classification and then on another basis is called
cross-classification. It can be repeated as many
times as available basis of classification.
Classification of data can be compared to that of
sorting letters in a post office.
 Objective –
The principal objectives of classifying the data are –
To condense the mass of data in such a manner that
similarities and dissimilarities can be readily
apprehended. Millions of figures can thus be arranged
in a few classes having common features.
To facilitate comparison.
To pinpoint the most significant features of the data
at a glance.
To give prominence to the important information
gathered while dropping out the unnecessary
elements.
To enable a statistical treatment of the data collected.
Types of classification –
It can be classified on the basis of following four
criteria –
Geographical, i.e., area-wise e.g. cities, districts,
etc.
Chronological, i.e., on the basis of time.
Qualitative, i.e., according to some attribute.
Quantitative, i.e., in terms of magnitude.
Geographical Classification – Here the
classification is done on the basis of geographical
or vocational differences between the various
items like countries, states, cities, regions, zones,
areas etc. Geographical classification is usually
listed in alphabetical order for easy reference.
Items may also be listed by size to emphasis the
important areas.
Chronological Classification – When data are
observed over a period of time the type of
classification is known as chronological
classification. E.g. figures of population.
Qualitative Classification – In this type of
classification data are classified on the basis of
some attribute or quality such as gender, color,
literacy, religion, etc. Here the attribute can not
be measured but can only be found out.
e.g. population as rural / urban.

Quantitative Classification – It refers to the

classification of data according to some
characteristics that can be measured. E.g. height,
weight, income, sales, profits, production, etc.
Frequency distribution is
Tabulation of Data –
One of the most revealing and simplest way for
summarizing data and presenting them in a
systematical way is the statistical table. It is a
systematic arrangement of statistical data in rows
and columns. Rows are the horizontal
arrangements while columns are the vertical
ones. The table is created to simplify the
presentation and facilitate comparisons. The
reader is enable to read the data quickly and
assess it.
Difference between Classification and Tabulation
classification is the first step in tabulation. It
means that before we tabulate the data it has to
be classified first i.e. different items with similar
characteristics must be brought together. After
classifying the data it is displayed under various
rows and columns for further simplification.
Role of Tabulation –
It simplifies complex data – When data are
tabulated all unnecessary details are avoided.
Data are presented systematically in columns and
rows. Hence the reader gets a very clear idea
about the table contents.
It facilitates comparison – As a table is divided
into various parts and for each part there are
totals and sub-totals, the relationship between
different parts of data can be studied much more
easily with the help of a table than without it.
It gives identity to the data – when data are
arranged in a tabular form with a title and
number they can be distinctly identified and can
be used as reference during problem solving.
It reveals patterns – Tabulation reveals patterns
within the numbers which cannot be seen in the
narrative form. It also facilitates the summation
of the figures incase the user wishes to check the
totals.
Parts of a table –
Table number – Each table should be numbered. It
helps in identifying or referring it during future
needs.
Title of the Table – it should have a precise title
describing its content, location and time.
Caption – It refers to the column headings. It
represents the contents of the column under it.
Caption should be clearly defined and place at the
middle of the column.
Stub – Stubs are designations of the rows or row
headings. They are to extreme left and its function
is similar to that of caption.
Body – The body of the table contains the
numerical information. It is the most important
part of the table. Data is arranged as per the
caption and stub.
Headnote – It is a brief explanatory statement
applying to all or a major part of the material in
the table and is placed below the point centered
and enclosed in brackets. It is used to explain
certain points relating to the whole table that
have not been included in the title nor in the
captions or stubs.
Footnotes – Anything in a table which the reader
may find difficult to understand from the title,
caption and stub should be explained in
footnotes. If footnotes are needed they are
placed directly below the body of the table. They
are used for following –
1. To point out any exceptions.
2. Any special circumstances affecting data.
3. To clarify anything in the table.
4. To provide source in case of secondary
data.
Rules of Tabulation –
a. It should suit the size of the paper usually with
more rows than columns.
b. In all tables the captions and stubs should be
arranged in some systematic order. It becomes
easier to read and understand the table.
c. The point of measurement should clearly be
defined and given in the table such as income in
rupees or weight in pounds.
d. Figures should be rounded off to avoid
unnecessary details in the table and a footnote
to this effect should be given.
e. If certain figures are to be emphasized then
they should be enclosed within a box or a circle.
f. The table should not be overloaded with
details. If many characteristics are to be shown it
is not necessary to load them all in one table. In
such cases separate tables should be prepared.
g. A column entitled miscellaneous should be
added for data which do fit as per the
classification.
h. The arrangements of the table should be
logical and items related to each other should be
placed near about and if possible in the same
group.
 Diagrammatic and Graphical Presentation
Data can be represented in the form of diagrams
and, graphs and maps. The problem lies in
selecting an appropriate way to present data. It is
a difficult task as it requires a lot of imagination
and talent on the individual or agency engaged in
the preparation of diagrams and graphs.

Significance of Diagrams and Graphs

Diagrams and graphs are extremely useful
because
 Significance of Diagrams and Graphs
Diagrams and graphs are extremely useful
because of the following reasons –
a. It furnishes an elaborated view of the data
which is easy to understand.
b. It looks attractive to the eye and create
interest.
c. It adds to the memory i.e. the impression
created by a diagram last longer than those
created by the figures.
d. It facilitates comparison of data relating to
different periods of time of different regions.