Marketing Research Tool

CONTENTS
Unit Page
TITLE
No. No.
1. Fundamentals of Market Research

1.1 The Marketing Environment
1.2 Market Research
1.3 The Need for Market Research
1.4 Applications of Market Research
1.5 Threats to Market Research
1.6 Market Information System (MIS)
1.7 Market Decision Support System (MDSS)
1.8 Summary
1.9 Keywords
1.10 Case Study
1.11 Self-assessment Questions
Application and Limitations 19

2.1 Introduction
2.2 Research Activities in Large Companies
2.3 The Client/Researcher Interface
2.4 Researcher Obligations to a Client
2.5 Decision Making and Research Inputs
2.6 Limitations of Market Research
2.7 Summary
2.8 Key Words
2.9 Case Study
2.10 Serf-assessment
Questions
39
Sampling
3.1 Introduction
3.2 What is Sampling?
Unit Page
No. TITLE No.
3.3 Reasons for Sampling
3.4 Estimating and Testing of Hypothesis
3.5 Definition and Managerial Objective of Sampling
3.6 Problems Associated with Sampling
3.7 Non-response Issues in Sampling
3.8 Factors Determining the Sample Size
3.9 Advantages of Sampling
3.10 The Sampling Process
3.11 Summary
3.12 Keywords
3.13 Case Study
3.14 Serf-assessment Questions
4. Questionnaire - Design Development 61

4.1 Introduction
4.2 Functions and Classification of a Questionnaire
4.3 Flow Chart for Questionnaire Design
4.4 Types of Questions
4.5 Choice of Survey Methods
4.6 Close Ended and Open Ended Questions
4.7 Sample Questionnaire
4.8 Summary
4.9 Keywords
5. Scaling Methods 97
5.1 Introduction
5.2 Types of Scales
5.3 The Concept of Attitude
5.4 Criteria for a Good Test
5.5 Forms of a Rating Scale
Unit Page
No. TITLE No.
5.6 Selecting the Appropriate Scale to Use
5.7 Summary
5.8 Keywords
Data Collection Primary / Secondary 117

6.1 Introduction
6.2 Quality of Data and Data Types
6.3 Data Types
6.4 Sources of Data
6.5 Communication Media
6.6 Formulating a Data Acquisition Plan
6.7 Basic Methods of Gathering Primary Data
6.8 Advantages and Disadvantages of Primary Data
6.9 Types of Secondary Data Collection and its
Advantages and Disadvantages
6.10 Respondents and their Descriptors
6.11 Summary
6.12 Keywords
6.13 Case study
Data Preparation and Processing 139

7.1 Introduction
7.2 Traditional Approach of Data Processing
7.3 Principles of Coding
7.4 Editing
7.5 Tabulation
7.6 Summary
7.7 Key Words
Unit Page
_No._ TITLE No.
8. Data Analysis and Interpretation 159

8.1 Introduction
8.2 Nature of Interpretation
8.3 Overview of Data Analysis Process
8.4 Statistics Associated with Frequency Distribution
8.5 Measures of Shape
8.6 Measures of Dispersion
8.7 Summary
8.8 Keywords
9. Multivariate Analysis 173
9.1 Introduction
9.2 ConjointAnalysis
9.3 Applications and Problems of Conjoint Analysis
9.4 FactorAnalysis
9.5 Model, Application and Limitations of Factor Analysis
9.6 Cluster Analysis
9.7 Multi-dimensional Scaling (MDS)
9.8 Summary
9.9 Key Words
10. Research Process and Design 193

10.1 Introduction
10.2 Research Design
10.3 Formulating the Research Problem
10.4 Features of a Good Research Design
10.5 Classification of Research Design
10.6 Summary
Unit Page
No. TITLE No.
10.7 Keywords
10.8 Case Study
11. Specialised Techniques in Market Research 209

11.1 Introduction
11.2 Types of Specialised Techniques
11.3 Research in Advertising Decisions
11.4 Motivation Research
11.5 Techniques of Motivational Research
11.6 Limitations of Motivational Research
11.7 Research for Pharmaceutical Products
11.8 The Source of Information for Pharma Products
11.9 Research for Industrial Products
11.10 Sources of Information for Industrial Products
11.11 Summary
11.12 Key Words
12. Market and Sales Analysis Research 231

12.1 Introduction
12.2 Types of Sales Analysis
12.3 Areas of Sales Analysis
12.4 Methods for Measuring Market Potential
12.5 Sales Forecasting
12.6 Methods of Sales Forecasting
12.7 Summary
12.8 Keywords
13. New Product Development & Test Marketing 249
13.1 Introduction
13.2 New Product Development
13.3 New Product Development Process
13.4 Test Marketing
13.5 Uses of Test Marketing
13.6 Guidelines for Test Marketing
13.7 Advertising Research
13.8 Importance of Advertising
13.9 Need for Advertising Research
13.10 Media Research
13.11 Copy Testing
13.12 Examples of Advertising Research Studies in India
13.13 Evaluation of Advertising Research
13.14 Summary
13.15 Keywords
14. Presentation of Research Findings 279

14.1 Introduction
i
14.2 Types of Reports
14.3 Report Content
14.4 Principles of Report Writing
14.5 Various Graphic Representations and Ethical Issues
14.6 Managing Market Research
14.7 Summary
14.8 Keywords
Unit Page
No. TITLE No.
15. Research in Service Segments 299
15.1 Introduction
15.2 Research through Internet Era
15.3 Methods of Online Research/Surveys
15.4 Disadvantages of Online Surveys
15.5 Ethical Issues in Marketing Research
15.6 Responsibility towards a Client
15.7 Summary
15.8 Keywords
Marketing Research
1.1 THE MARKETING ENVIRONMENT
I have a good product, I have a good brand name, I have a good marketing network and
I give the best price. I have goodwill from my customers - but still I am not achieving the
volume that I want. What is wrong? Is my product not appreciated by the customer or
haven't I reached my target customers? Does my price bother the customer while he / she
is buying? To find the solution for all these problems, I need to do a systematic research of
all the above activities. Today these questions are asked by every organisation, but only a
few organisations act on the findings of their research and they become more successful
with the right strategies. The organisations which do not act on time fail to sustain their hold
on the market.
Today, in the competitive market, one has to study all the aspects of the marketing mix
-viz. the Product, the Place, the Pricing and the Promotional/Advertisement activities, if
one wants to become successful in the market for a long time. Then all the above parameters
are to be studied carefully and then launch/re-launch the product with the right price to the
right customer (target customer). For that, a detailed research has to be conducted - not
once but on regular intervals so as to be in the market for a long time, cate to the needs of
the customers and satisfy the customers with profits to concerned organisation.
1.2 MARKET RESEARCH
Before going on to Market research, let us understand what is meant by the term "research".
Research always starts with a question or a problem. The purpose of research is to find
answers to questions through the application of scientific method. It is a systematic and
intensive study directed towards a more complete knowledge of the subject method.
The term marketing is a broader concept which includes all activities, which are related to
the transfer of ownership of goods and service e.g. product development, packing, branding,
warehousing, distribution, advertising, sales promotion, sales service and so on.
Market is a narrow concept which refers to all potential customers sharing a particular
need/want and might be willing to engage in activities to satisfy those wants.
Research is a process of gathering, recording and the analysis of relevant and critical
facts about a specific problem. Thus market research as per the American Marketing,
Association is "The systematic gathering, recording and analysing of data about
problems relating to the marketing of goods and services".
Another simpler definition can be put in this way.
I
Unit 1 Fundamentals of Market Research
Market Research is the function which links the seller and the buyer through information
-information used to identify and define marketing opportunities and problems, generate,
evaluate the actions and monitor marketing performance by improving and understanding of
the market as a process. Key aspects of the definition
• Market research is a problem-solving device.
• It is a continuous and evolving process as the needs and wants keep on changing.
• It is a well-defined, planned process comprising of systematic and intensive
investigations.
• It adopts scientific methods and objectivity in solving marketing problems.
Research can be classified into two broad categories:
i) Basic research ii) Applied research

Basic research is sometimes called fundamental research, theoretical research or 'pure'
research. It aims at expanding the frontiers of knowledge and does not directly involve
pragmatic problems. The essence of basic research is that it addresses itself to more
fundamental questions and not to the problems with immediate commercial potential.
Applied research, which is also called decisional research, on the other hand, proceeds
with a certain problem, and it specifies alternative solutions and the possible outcomes of
each alternative. Unlike basic research, it is prompted by commercial considerations. Though
one may usually be able to distinguish between basic research and applied research, the
distinction between the two sometimes gets blurred. Several firms may be engaged in
basic research which does not have any immediate commercial use. However, it may be
potentially commercial or else the firms would not have undertaken it at all.
Applied research can be divided into two categories:

i) Problem-solving research
ii) Problem-oriented research
Problem-solving research, as the name implies, is concerned with a particular issue or a
problem and is usually proprietary in character. The latter characteristic indicates that such
a research is undertaken within a firm or by an outside consultant on its behalf.
Marketing Research
Problem-oriented research, on the other hand, is concerned with a class of issues or

problems in which several firms may be interested. Research of this type is usually concerned
with the conceptual aspects but is oriented towards applied problems. ,
i
x^T Activity A;
a) What do you mean by the term'research'?
b) Do you think that market research is undertaken to guide managers in the analysis of
marketing problems? Critically examine this statement. List down any two major
problems in marketing.
1.3 THE NEED FOR MARKET RESEARCH

Market research is not a substitute for top management judgment but is an aid to judgment.
For any organisation, large or small, profit or non-profit institution, a systematic study of
their market would be useful in decision-making. The market research function in various
government departments will be useful to entrepreneurs and other business organisations.
They will be able to get authentic information on public policy impact, economic
development, technology transfers, and planning. Market research is the foundation stone
for marketing concept. Managers are separated from their customers. They need
information on their consumers. The consumer needs can be studied and thus goods and
services can be produced as desired by the consumers for greater satisfaction and
achievement of organisational goals. The significance of market research is increasing due
to the shift from local, national to international marketing and from price to non-price
competition.
There are several factors which have contributed to the research:

Marketing Research
run was approx 20,000 copies. Indications are that within a short time it has become one
of the leading monthly popular marathi magazine.
Nestle India Limited launched Small KTTKAT at Rs 5/- from its international portfolio. It
is wrapped in vibrant colours of White and Red in conformity with its international packaging.
Earlier, Nestle India Limited commissioned a consumer research study. The results of this
study were encouraging and showed that the Indian customers is always interested in
experimenting with new food options viz small quantity (Chotta pack) and at less price.
Pepsi Foods has assigned great importance to Market research. Through research it gets
systematic information about its markets and its customers. All its research is done by the
IMRB. Broadly, research studies done for Pepsi Foods fall in the following three areas:
(i) Studies undertaken on a continuous basis like marketing tracking studies and
retail audits.
(ii) Studies that are commissioned for specific marketing problems faced by the company.
(iii) Studies done from time to time as per the requirement of the company such as a
study to ascertain the effectiveness of an ad campaign.
All these three types of research studies have tremendously helped Pepsi Foods to strengthen
its position in the market. It feels the pulse of the market and is always in touch with the
latest developments in the market.
Multinational company Whirlpool Asia lays considerable emphasis on Market research.

In this company, every activity, strategy and decision is based on data collected through
the research process. It believes in planning research in advance though it is rather difficult.
It strives to have a meaningful dialogue with the consumer in order to know his real opinion
about its products, what difficulties he experiences and what suggestions he has to offer.
Information thus received proves to be quite useful to the company in modifying its products
or in evolving new ones.
Whirlpool has gained an insight into the various segments in the market. In India, it has
segmented the market on the basis of the different stages of the product life cycle. Now
they have assigned their brand ambassador married couple as Kajol and Ajay to advertise
their product.
Decisions like which size of refrigerator should be put in the market or what should be the
price of a particular model are based on research. Marketing and Research Group (MARG)
has been the main Market research agency for Whirlpool.
JS$ Activity B:
a) Do you really think that market research should be done on a regular basis by every
organisation? Can you mention any organisations where you feel that this happens on
regular basis?
b) In your opinion, where would be the problems in the different areas of marketing, in
which market research can be helpful? Give details of any four problems which you
have seen.
A. Market research
As market research does not address itself to basic or fundamental questions, it does
not qualify as basic research. On the contrary, it tackles problems which seem to
have an immediate commercial potential. In view of this major consideration, market
research should be regarded as applied research. We may also say that market research
is of both types—problem-solving and problem-oriented.
Market research is a systematic and objective study of problems pertaining to the

marketing of goods and services. It may be emphasised that it is not restricted to any
particular area of marketing, but is applicable to all its phases and aspects.
In short, market research provides the requisite information for making marketing
decisions.
B. The Growth of Market research

Over the past several decades, there has been a gradual development in the field of
market research. The first evidence of formal research techniques being employed
could be traced as far back as 1824.
Marketing Research
However, for almost one century thereafter, market research developed rather slowly
and sporadically. It was only from 1910 that it made rapid strides. Several factors
have contributed to the growth of market research.
First, as a result of large scale production, producers do not have direct contact with
the consumers. This leads to problems in the marketing of goods. It is necessary for
a manufacturer to know the potential areas where his goods could be marketed, the
probable demand for the same and the extent of profitability and competition. This is
possible only when a systematic investigation is undertaken.
Second, there has been a shift from a seller's to a buyer's market. This necessitates
a better understanding of buyer behaviour. This aspect is studied in market research.
Third, the emergence of specialists such as statisticians, psychologists and behavioural

scientists in a fairly large number has also enhanced the importance of market research
in no small measure. These specialists have helped in the development of market
research as a separate discipline and strengthened its techniques.
Fourth, the increasing use of computers. The IT sector has further contributed to the
growth of market research. Finally, the changes in the composition of population,
particularly the shift from the rural to urban areas has given a wide scope for marketing
of various types of goods and services in an urban area.
C. Market research for Industrial Goods
Let us know the difference between consumer market research and industrial market
research. This difference arises on account of the distinction between the markets for
consumer and industrial goods. The following are the main points of distinction between
the two:
1. Derived demand
One way of classification of demand is: autonomous (direct) and derived (indirect)
demand. The demands for industrial goods are derived demands as they are
required to obtain consumer or industrial goods. For example, the demand for
compressors to be used in the manufacture of refrigerators which are consumer
goods, is a derived demand. Another example could be the demand for steel
which is required for the manufacture ofjrmghings, which are industrial goods.
Thus, the derived demand is aifmdirect demandjthere is a chain of such a
derived demand. The larger the chain, the morecomplicated is the process of
analysis compared to the analysis of the autonomous or direct demand for
consumer goods.
2. Demand concentration
There are three types of concentrated demand for industrial goods:
• Geographic
• Industrial
• Purchasing
Geographic concentration implies that there is a general tendency among many

industries to cluster in selected areas. For example, this may be on account of
the proximity of the sources of raw material. Industrial concentration may be a
result of the inherent characteristics of industrial goods and services, leading to
a limited number of such markets. Purchasing concentration results from the fact
that a few firms account for a relatively large proportion of the total demand for
industrial goods.
Unlike industrial goods, consumer goods do not have these three types of
concentration in demand. There is no geographic concentration as consumers
are scattered all over the country. There is no industrial concentration as markets
for consumer goods are not confined to limited locations. Finally, there is no
purchasing concentration as consumer goods are not purchased in bulk by a
few consumers. This is because the demand for consumer goods emanates from
every consumer who buys in a limited quantity as per his requirement.
3. Demand volatility
The demand for industrial goods is subject to sharp variations at times. This
happens because of factors such as drastic changes in business conditions,
financial considerations and price. Industrial buyers react to such changes by
varying the level of inventory accumulation and advancing or postponing the
timing of purchases of industrial goods. In contrast to the demand for industrial
goods, the demand for consumer goods is not subject to sharp fluctuations,
particularly in respect of items of everyday use.
Consumers have to buy their daily requirements even though the prices may be
higher. Broadly, the demand for consumer goods would be less volatile on
account of changes in the business environment and financial stringency than
that for industrial goods.
Marketing Research
4. Purchasing procedures and organisation '

As there are several buying influences affecting industrial firms, generally an
industrial purchasing decision involves more persons than a consumer purchasing
decision. In view of this, a study of attitudes, motivations and the relative influence
of purchasing decisions is more complex than that for consumer goods. On the
other hand, persons deciding to make industrial purchases are better informed
than the average consumer. i
5. Manufacturer-buyer relationships
The relationship between the manufacturer and buyer of industrial goods is

stronger, closer and more continuous than that for consumer goods and services.
This is primarily because there are less intermediaries for industrial goods than
for consumer goods.
Whenever market research relates to industrial goods, these differences between industrial
and consumer goods markets must be borne in mind. Suitable changes in the emphasis of
market research techniques should be introduced. To apply the same techniques,
disregarding the substantial differences between industrial and consumer goods markets,
would undermine the practical utility of market research for industrial goods.
Activity C;
a) Many times the management is not convinced about the money spent on market
research as they think of it as an unnecessary activity over which no funds should be
spent. Do you agree to this statement?
b) The field of market research is very large as it covers a wide variety of problems.
List any five problems which might be applicable.
10
1.5 THREATS TO MARKET RESEARCH

The success of Market research depends upon the cooperation of the respondents,
unprofessional or unethical Market research may seriously damage the research which in
turn will give bad results in the reports the researchers may also draw wrong judgements
and results which will hamper in the decision making of their client. It is necessary to
ascertain the reaction of the respondents, the three major issues which causes threats to
marketing and survey research are:
• Excessive interviewing
• Lack of knowledge
• Amusement of respondents
t The use of Market research as a sales ploy
It is noted that these factors pose a threat to Market research especially in the Developed
and western countries. In our country Market research is still used very rarely so these
threats are almost non-existent.
Researchers carried out reappraisal of these threats. They reported the results from five
cross sectional studies conducted at two intervals the findings were as follows;
They investigated three areas excessive interviewing lack of knowledge and abusement of
respondent and use of market research as sales ploy, it was found that use of market
research as sales ploy was the greatest threats if it is not checked it is likely to erode the
trust of customer respondents, respondents will be suspicious of the interviews and will
not cooperate in the right manner.
The researchers have expressed their concern over the fact that almost forty percent of
the college educated, middle upper income respondents and more than half of the
respondents living in the western region of U.S. have experienced Market research as
sales ploy. They feel that this may hamper the actual genuine people conducting the research,
particularly as these people may be opinion leaders of their better education and affluence.
The other two threats were not as serious as compared to the third one but there is a
potential danger in these fields also, because as research depends upon the respondents
cooperation its depletion will adversely affect the interest of Market researchers. It was
also concluded that both excessive interviewing and lack of consideration and abuse of
respondent should be avoided.
Marketing Research
This can be done by screening the questions so that the persons who have been interviewed
recently or too frequently can be eliminated. It is also advisable to use non-interviewing
research techniques wherever possible so that the need to approach the respondents is
minimized
1.6 MARKET INFORMATION SYSTEM (MIS)
Market research is said to be different from Marketing Information System (MIS). Let us
define MIS as a structured, interacting complex of persons, machines, and procedures
designed to generate a flow of pertinent information collected from internal and external
sources for the use as a base for decision making in specified areas of marketing
management.
The definition indicates the interdependent activities associated with the collection of
marketing information from internal and external sources. It also shows that such information
is collected to draw the conclusion needed for decision making in different areas of
marketing management. Market research is a means of obtaining information to be used
for marketing decisions. If compared, both the concepts show that while market research
generates information, MIS concentrates more on the storage and flow of information to
marketing managers. This indicates that marketing information system is a much wider
concept than market research. Good MIS determines the information needs of the
organisation and generates and processes such information on a continuous basis. It should
also provide for its storage so that it can be used as and when required.
Let us analyse what type of information is required for our MIS systems. There are three
types of information required for marketing managers, they are;
1. Recurrent information
2. Monitoring information
3. Requested information
Recurrent information: This type of information is provided on periodic basis, which

pertains to
• information on sales
• market share
• customer satisfaction
• perception of customers
• advertising expenditure
These are or may be supplied on a weekly or monthly basis.
Monitoring information: This type of information is obtained from regular scanning of

certain sources. The sources are:
• Official publications
• Journals
• Annual reports
These sources can be very helpful to organisations as they indicate the nature of problems
that are likely to arise and the possible changes in the business environment, these can be
also helpful in identifying new market segments, the new use of existing products and also
the possibilities of improving product by introducing new features.
Requested information: Requested information as the name suggests, this is sought by

marketing managers. This can involve a wide range of activities such as
• cost and price analysis of a competitive product
• cash flow of competitive companies
• quality testing of competitive companies
These are few to name, such type of information would not be usually available unless a
request is made for the same. Once a request is made for a specific information, then a
series can be built up over time provided that such information is needed by the managers.
1.7 MARKET DECISION SUPPORT SYSTEM (MDSS) __________

In a highly competitive environment, especially in the western countries, marketing managers
find themselves seeking a wide range of information in order to take the right decisions.
This means that they should not only be concerned with the availability of the requisite
information but also with its accuracy and adequacy. This paved the way for the creation
of a marketing decision support system (MDSS). This system consists of a collection of
data, storage, analysis, and the reporting of marketing data. The new generation of
computers and advanced software has given a boon for setting up the MDSS.
13
Marketing Research
The main difference between MIS and MDSS is that whereas the former is centralised,
the latter is decentralised and allows marketing managers to interact directly with the
database.
MDS S can be classified into five components

1. Data Sources "
2. Data Management
3. Display
4. Statistical analysis
5. Modelling
Data sources are major components of MDSS. These sources can be both internal and
external. For example, a company's accounting system is an internal data source. This
source provides data pertaining to orders, sales, inventory levels, receivables and payables.
Library sources as well as syndicate services are the examples of external data sources.
Library sources contain a wide range of information such as Government publication
reports, sales of periodicals, buyer profiles, sales of retail outlets, evaluation of advertising
campaigns, major economic, social and demographic trends, etc. Both, library sources
and syndicate services can be immensely useful to the company.
Owing to a variety of internal and external data sources, it becomes necessary for the
company to device a proper database management. This involves identifying relevant data
from a variety of sources, deciding on the extent of data and keeping the data in an easily
available form.
Data formats allow the retrieval of data easily and promptly. Display formats could range
from simple adhoc retrieval to more formal displays.
A company may have a graphic display package consisting of such options as time-series
plots, bar and pie charts and scatter diagrams where the comparison of two measures is
involved.
Statistical analysis involves the performance of different statistical operations such as the
calculation of averages, standard deviation and regression. Such calculations are helpful in
understanding the relationship among marketing variables. This, in turn, brings out the
underlying' marketing issues in sharp focus.
14
Modelling is the last component of MDSS. Models are mathematical formulations.

Marketing decisions can be tested against the model to ascertain what could be the possible
outcome in a given situation.
£$ Activity D;
"Market research is undertaken to guide managers in the analysis of marketing problems."

Critically examine this statement and comment.
1.8 SUMMARY _________________________________________ _ _ _ _

Market research should be given prime importance in every organisation. This not only
helps us to analyse our weakness but also gives first hand information about the latest
trends going in the market. One has to learn from the reports and take appropriate actions
on which strategies are to be adopted in the near future. Research should not be taken in
a negative way but one has to learn from the mistakes and adopt the appropriate way for
successful marketing. The choice of research activities in the organisation gives a solution
to the exact method to be adopted for survey.
1.9 KEYWORDS _____________________ _ _ _ ^ _ _____________

Authentic: of undisputed origin, or original
Entrepreneur: a person who undertakes a commercial venture

Emphasise: stress
Inherent: existing in something as an essential or permanent attribute
Pertinent: relevant
Perception: act or faulty of perceiving
15
1.10 CASE STUDY
Sony India
Sony, a famous global brand, stepped into India in 1995, in the post liberalisation period.
Within a couple of years, it started its manufacturing unit. The company makes its presence
through Sony Entertainment Television, Sony Music and Sony India. The first two are
comfortably placed while the third is expected to bring all the digital products.
The plan was cleared by the government, but Sony India is facing problems, one of the
major problems is the grey market. While Sony had targeted their sales around 1000
crores by the end of the century, it has to have a good understanding of India's economic,
social and political environment.
Sony India likes to develop itself as its parent company developed business units in US,
Europe, and Asia (Singapore), yet it is a challenging task for them in India.
Questions
a) Will research, play an important role in the development of Sony India?
b) Which will be the best research - hiring an outside agency or in-house research ?
c) For considering an outside agency what factors should it consider while selecting the
agency?
16
Marketing Research
2.1 INTRODUCTION
With the increasing use of market research these days, it is being realised that market
research is a complex task and has to be properly managed if optimum results are to be
obtained. At the outset, the question arises as to what we understand by research
management.
Research management concentrates mainly on the processes, projects, personnel, finances

and agencies engaged in research. Its prime responsibilities include monitoring of the research
design, for which the study is carried out properly in line of the management objective. It
also monitors that the presentation of the result to management is handled properly and
that all these activities are administered within a controlled budget. It means that proper
research organisation has to be set up to handle individual projects and task.
It will be seen that research management includes a host of interrelated activities. The
types of research projects to be undertaken, the selection of research personnel, financing
of research manager should be the concerns of a research management. A research manager
has to ensure that a suitable research design is developed and that the study is carried out
on the right lines, according to a predetermined schedule and within the budget allocated
for the purpose. As these activities are interrelated, if any one of them is not conducted
properly, it will have an adverse impact on the other. For example, if a research project is
not conducted according to the time schedule laid down, it will increase the cost and make
it impossible to complete the project within the allocated funds and the time span.
A. Importance of research management

Just as the marketing concept is useful in attaining the objective of integrated marketing,
research management is relevant for making market research really useful. By superior
management, various disjointed research techniques can be pulled together and
coordinated so that market research can be made really effective.
The marketing managers or the research managers who understand only techniques
will be a failure such a study. They must understand how to put together all the efforts
falling within his sphere of responsibility. No longer is market research an isolated,
ivory-tower operation. It is an integrated whole, making a major contribution to
marketing in its own right.
An improved market research management firm can enjoy two major advantages,
viz:
(i) Market research will be more helpful to marketing management in
decision-making.
20
Unit 2 Application and Limitations
(ii) Market research can be conducted more economically and more effectively.
B. Qualities of a market research manager
The success of market research management depends to a large extent on the quality
of leadership. A research manager must have some special qualities if he has to do
justice to his job. Some qualities which he should possess will be
• innovative and always be prepared to look at any problem from a fresh

perspective.
• confident of his skill and knowledge and prepared not only to accept new ideas
but also to try them out.
• well-informed and up-to-date in his own area of specialisation.
• familiar with recent developments in particular industries, especially the problems

faced by them in the marketing of their products.
• familiar with the recent trends in research techniques and should be prepared to
absorb the latest knowledge as mush as possible.
• he should have considerable selling skills so that he can convince the management
regarding the accuracy and soundness of the recommendations emerging from
his research.
One of the most crucial decisions to be taken by management is whether market

research should be undertaken and if so, what place is to be assigned to this function.
This is an administrative decision which calls for a clear perspective on the part of top
management. Even when a decision is taken in favour of the market research function,
a related issue is what budget allocations are to be made to carry out market research
efficiently. The organisation of market research activity also has to be considered so
that management gets the maximum benefit from it. These are some of the main issues
which must be given due consideration by the management. Some of these functions
are discussed briefly below.
C. Organising market research function
First of all, the question arises as to how the market research function is to be organised
within a firm. It should be noted that the organisational structure for this function will
differ according to the type of agency which is handling it. Thus, a market research
organisation in a manufacturing firm will be very different from that in a research or a
21
Marketing Research
consultancy firm, as the requirements and emphasis on research will be different. A

major difference between a manufacturing firm and a research firm is that whereas in
the former, research being one of the several activities is a means to an end, in the
latter, research is the only activity and is an end in itself.
The research firm has to ensure that research must ultimately lead to profit otherwise
it cannot survive. In contrast, a manufacturing firm does not aim at profit although it
would expect that research will indirectly lead to increased profitability.
Another point of difference between the two organisations is the degree of emphasis
on the practical utility of market research. A manufacturer's firm would normally
place greater emphasis on the market research findings in the day-to-day marketing
of its products. But this need not be the case in a research firm as it is not engaged in
marketing of any products.
Coming to the organisation of research activity, several alternatives are open to a

firm. For example, there could be a market research department or cell with a skeleton
staff if the market research activity is not very important. Alternatively, the firm may
have a well-developed department comprising a number of specialists as also the
other supporting staff.
These are the two extremes within which the market research organisation would
vary. The specific organisation will depend on the requirement of each firm. It is
worth emphasising that a firm, intending to set up an organisation to look after market
research function, may not be in a position to set up the most suitable one in the very
first attempt. As its management acquires experience in managing the research function,
it goes on making improvements in its organisation. In course of time, such an approach
will enable the firm to develop the most suitable organisation to manage its research
function.
A firm which has decided to introduce a market research function has to decide
which of the three options that are normally available, should be selected:
a. to set up a formalised market research department consisting of some full-time

specialists whose main responsibility would be to carry out market research
studies and report the findings to the management.
b. to assign the market research responsibility to one or more line or staff executives
on a part-time basis, who would have to undertake it in addition to the major
responsibility.
22
I
c. to assign the responsibility of undertaking market research to an outside individual

or an organisation.
There are market research consultants who do research on behalf of the firm and are
paid their consultancy fees for the same. Small firms prefer to adopt the second or
third approach whereas medium and large-sized firms, which may have frequent
marketing problems, prefer the first one. The latter would also involve the maximum
financial commitment.
In some firms, where a separate marketing information system exists, the market
research function may form a part of such a system. In such cases, market research
acts as one of the inputs to the marketing information system. Wherever market
research is undertaken frequently, it is desirable to formalise this function and to
integrate it not only with marketing activity but with the overall management of the
organisation.
It should be noted that if a firm decides to have a separate market research department,
then sufficient freedom should be given to them to carry out its tasks. Many a time the
departments are set up with high hopes and expectations but are not given the
necessary finances and freedom. As a result, they are unable to do much for their
firms. Market research departments should be encouraged to develop, carry out and
analyse their research as they want. The top management must delegate necessary
authority to the research personnel to enable them to carry out their work smoothly.
Another aspect that is relevant is whether a large firm should have a centralised
market research department or whether alternatively each division or operating unit
should have its own research cell. The main advantage of the centralised market
research department is that there will be more effective coordination of market research
with marketing management. Effective control and supervision of research can also
be ensured. On the other hand, if market research is undertaken by each division or
operating unit, it will enable researchers to be more knowledgeable about divisional
markets and their problems. In addition, such an approach will offer greater autonomy
to divisions and operating units in their research activity which will ultimately be
beneficial to the firm.
Both these approaches are in vogue and it is difficult to say which is more popular. In
this context, it is pertinent to note that the organisation of market research function
need not remain static for all time to come. It should be dynamic and flexible so that
it can be adapted to the changing requirements of the firm.
23
Marketing Research
If a company has decided to set up a separate market research department, it is very

essential that it should be clear about its role in the overall organisation. The market
research department should take up problems faced by the company, study them
objectively and report its findings to the management. It should develop a plan, without
which it will lack focus and utility to the management. Like any other plan, the key
elements of a market research plan are the same, namely, objectives, policies, specific
projects, resources required both in terms of manpower and budget and a time schedule
for research projects. In identifying research projects, it is expected that clear priorities
are laid down on the basis of which they can be taken up for investigation. These
priorities should be matched by proper allocation of funds and manpower.
The market research plan should be written out in sufficient details. Before it is finalised,
it should be circulated amongst the concerned officials in the company. It should fit
into the marketing plan of the company as it is an important input in it. Although it
should be followed and implemented as far as possible, there should be an element of
flexibility in it so that certain changes, if found necessary, can be made.
Âctivity A:
a) List any five qualities of a good research manager.
b) Why is research regarded as useful to marketing management?
D. Selection of market research
Many a time, companies do not set up market research departments as they do not
have an adequate amount of research work to warrant a separate department. They
prefer to hire the services of a market research consultant as and when they feel the
need for research. In our country, the number of companies having their own market
24
research departments is extremely limited. Many 'buy' market research from outside.
Such companies should know how to select a market research agency before
sponsoring a research study.
E. Selection of research agency
Why should one hire market research agency? If a firm does not have persons or
well-versed market research department or cell, or has a market research dept. may
sometimes find at a particular point of time that, it is fully engaged on certain other
studies and, as such, it cannot take up an additional problem, for detailed investigation.
In that case, it has to assign the study to an outside market research agency. The firm
may find that an outside agency may complete the study faster, at a lower cost as well
as with the laid down objective. As such it may decide to assign its sudy to an outside
agency. Selections can be made from entities such as market research consultancy
firms, advertising agencies, suppliers of syndicated services, etc.
Now how does the firm go about choosing an outside market research agancy. In
developed countries, there are a large number of such agencies which either undertake
complete ad hoc research project or perform one or more specialised services such
as drawing a proper sample of respondents, interviewing respondents, or processing
data. In all those cases where it is necessary to hire outside research agancies, the
market researcher must be able to evaluate such specialised services. The criteria
for evaluating specialised agencies are as follows:
Criteria for hiring outside agencies

• Technical expertise
The market researcher should know who is to undertake the study and what is their
proficiency in market research. The client firm may find that a research agency is
good at basic studies but is not competent enough to undertake complex studies.
Some research agencies are poorly staffed and as such they should be avoided.
• Objectivity
The question of objectivity is very important. Outside agencies should be reputable
for their objective approach in research projects.
• Confidentiality
The client firm must ensure that the research agency maintains strict confidentiality
regarding the project/study.
Marketing Research
• Economic factors
A client firm may invite research proposals from more than one agency. In such a
case, it would choose the most economical agency. However, client firms should not
overlook the fact that some agencies are very economical because the quality of their
work is rather poor. It is not advisable to be economical at the cost of quality.
• Timely submission of reports
The client firm should enquire about the reputation of the research agency especially
in relation to its timely submission of reports. Sometimes, outside agencies are quick
in taking up assignments from clients but are not so prompt in carrying out the task.
• Experience of the supplier
The client firm should ascertain the standing of the agency. While general experience
is very important, relevant and specific research experience is what should be looked
for.
• Reputation of the agency
It is necessary to ensure that the agency has a good reputation. This consideration is
important for lending credibility to the research findings. This is of special importance
particularly when the client firm intends to use the study for creating an impact.
Since no single agency is likely to be strong on all these considerations, it is necessary that
the client firm adopt a reasonable approach in this regard. It should ascertain which of the
above criteria are crucial for its research project and then apply these criteria in selecting
an agency from amongst those short-listed. In order to facilitate comparison amongst the
agencies, the client firm should send the same study specifications to all of them for gathering
research proposals and cost estimates.
In our country, there are some organisations that take up research studies on behalf of
their clients. Acomplete list of such organisations, the nature of research usually undertaken
by them along with their strengths and limitations, is not yet available. Here, we are mainly
concerned with research projects undertaken by research agencies at the instance of their
clients. It may, however, be mentioned that some of these above mentioned organisations
are well established and have been engaged in syndicated research and related spheres of
activity for the past several years. They have professionally qualified staff, and they undertake
sponsored research on widely diversified themes.
26
Marketing Research
• Test marketing studies
• Corporate image studies
• Campaign evaluation - Media studies
• Opinion surveys
• Industrial market research
It may be mentioned that in our country, marketing research is still not well developed and
there are only a few large marketing research agencies in the country. However, some of
the studies carried out by these agencies were of pioneering nature, involving the development
of appropriate concepts as well as sampling and measurement techniques. They have a
large field set-up supported by full-time investigators in different parts of the country having
several years of professional expertise.
A company intending to buy marketing research may adopt one of the two courses. First,
it may straightaway ask a marketing research agency to prepare a research proposal
including an estimate of cost. Second, it may give some thought to its problem, spell it out
and provide certain specifications according to which the concerned agency has to carry
out the research. In this case the company will pay the agency its fees for carrying out the
task as per its specifications.
Advantages of research proposals
There are some advantages of getting a research proposal from an outside consulting firm.
There is conceptual thinking on the marketing problem referred to by the company. This is
because the outsider's approach is not subject to any constraints which may be applicable
within company.
It enables the company to evaluate the research capability of the consulting firm and its
dependability. When the company invites research proposals from more than one consulting
firm, it can compare them with respect to the nature and extent of work involved, the
quality of proposals, cost and time. This comparison will enable it to choose the most
appropriate proposal.
Research proposal is always desirable as it offers some sort of a commitment from the
consultant to the company. In its absence, there may arise some misunderstanding and
differences between the client and researcher regarding the specifications, time or price.
28
Last but not least, the research proposal once approved by the company, becomes a
contract, binding both the parties. The company can then turn to other problems, since it
is sure that the research will be taken care of by the consulting firm.
Contents of research proposal
It is very important to know the contents of a research proposal. While the style and
format will differ from one consulting firm to another, the research proposal should invariably
deal with some important aspects of research such as:
• A research proposal should give some background of the problem, indicating the
manner in which it is to be investigated. Then, a clear statement of the problem needs
to be given. Utmost clarity is required in defining the problem.
• A research proposal should specify the research methods to be used in the study.
This part should contain information on the sample design and the sample size, the
designing of the questionnaire, data collection procedure, and the processing and
analysis of data. This is the most important part of a research proposal.
t A proposal should then indicate the form and content of the final research report. It
should also indicate if the consulting firm will make a personal presentation of research
findings.
• A proposal should indicate the time needed to carry out the task and also the cost. It
is desirable that a broad time schedule covering major research operations be given.
The consulting firm should also state the assumptions on which cost and time estimates
have been worked out. A statement of this type will help avoid any misunderstanding
that might otherwise arise.
JS$ Activity B:
a) List out five qualities required for a market research agency.
29
Marketing Research
b) What criteria should be considered while hiring an outside agency? Name any
three.
c) List the important Qualitative and Consumer Research.
d) Could you list out on your own any new Qualitative Research which you have
noticed in your curriculum?
2.2 RESEARCH ACTIVITIES IN LARGE COMPANIES _________________
Earlier, we have read the different types of research taken by the agencies, some of the
common activities are listed by the type of research activities takenup by large companies
which they do by themselves or hire an agency to do the study.
Research activity *Usage frequency
Business economics and Corporate research
1. Short range forecast (<=1 year) High

2. Long range forecast High
3. Plant /Warehouse studies Average
4. Acquisition studies Average
30
5. Pricing studies High

6. International trade Low
7. MIS (Management Information System) Average

8. Product mix studies Low
9. Operation research Average
10. Company Employees Average
Sales and market research
1. Market Potential measurement High
2. Market share analysis High
3. Market characteristics High

4. Sales analysis High
5. Establishment of sales quota & territories Average
6. Physical Distribution alternatives Average

7. Test marketing Average
8. Consumer panel operations Low
9. Sales force effectivity studies Average

Advertising research
1. Motivation research Low

2. Copy research Low
3. Media research Average
4. Ad-effectiveness research High
5. Sales Promotion studies Average
6. CompetitiveAdvertising Average
Product/service research
1. New Product Mgh
Acceptance
3
1
Marketing Research
2. Testing existing Products High

3. Competitive Products High
4. Packaging Research High
5. Brand Loyalty Studies High
Corporate responsibility research

1. Ecological impact studies Low
2. Legal constraints (Advertising & Sales Promotion) Low
3. Socio-Political studies Low
*Frequency of usage
High 75%
Average 51 -74%
Low 50%
Note: The usage of research pattern may vary based on consumer goods, industrial
goods and services.
2.3 THE CLIENT/RESEARCHER INTERFACE ________________________

To get maximum benefit of research, the client and researcher must co-operate and
collaborate with each other in the following ways. The person in-charge must give full and
frank information on
1. Precise problem (What the problem is?)
2. Background and environment related to the problem (this is in regards to down in the
sales due to brand image or pricing problem)
3. Time and cost constraints (this is very important because the study is to be completed
in the stipulated time. For example, if one has to decrease the prices effectively from
the new financial year then the study is to be completed three months in advance
prior to the union budget.)
4. Limitations if any.
32
5. Data availability with the company (refers to the primary or secondary data)
6. Any situational change while the study is proceeded (may be the raw material prices
have suddenly gone up or may have dropped)
2.4 RESEARCHER OBLIGATIONS TO A CLIENT _________ ,

1. Honest and clear reporting (the reports should not be fake and should have
confidentiality)
2. Briefing regarding study limitations (what limitations have they faced while collecting
the data)
3. Explanation of report conclusions and aiding client's application (the conclusion should
be presented in a detailed presentation to the top management)
4. Demanding complete facts and expliciting the goals and problems faced (the problems
faced by the researcher e.g. some of the respondents have not given the correct
answers or may have given fake information).
5. Unwillingness to bow to client's biases, prejudices and pre-judgment.
2.5 DECISION MAKING AND RESEARCH INPUTS

The Research Inputs depend upon the problems faced by the organisation. These cannot
be same for each organisation for example, a firm has a good reputation in the market but
their product is priced high, hence, he will be interested in survey of price factor and a part
of that survey can be assigned for customer satisfaction in terms of quality and service of
the product. On the basis of the results, he has got he can take decisions and adopt the
right strategies for costing of his product or any additional services to be added in the
same cost.
2.6 LIMITATIONS OF MARKET RESEARCH________________________

1. Inexact science
Market research is concerned with human beings for information, the most dynamic
factor; hence probable trends can be identified. Moreover, in absence of proper
problem definition, it becomes a data collection exercise.
33
Marketing Research
2. Limitation of the researcher
Inexperienced, partially trained and educated researchers can hardly give quality
work. Besides this the element of subjectivity mars the findings because of individual
psyche and background.
3. Expensive exercise
Most of the organisations hire outside agencies to carry out extensive research, which
requires a lot of time and money.
4. Time lag
There is a difference of time between the research and its implementation. In the
mean time, the marketing dynamics may render it obsolete.
5. Historic data
Market research delves into the past data to predict the future, which has its own
limitations.
6. Interpretation of problems
Certain problems are complex in nature and various elements are involved which
cannot always be measured e.g., psychographic studies. Ascribing the result to specific
factors becomes difficult in these cases.
7. Environmental constraints
Market Research is likely to reflect the environmental constraints it functions under.
8. Fragmented v/s Holistic approach
Due to financial, time and management constraints partial problems are studied which
do not have a great impact, as it would have been with a holistic approach.
9. Inadequate commitment
Often organisations share partial facts and commitment towards research. Under
isolated condition the results are erroneous. Also it is used as a window dressing tool
to justify decisions already taken. (Pseudo research)
34
Marketing Research
Ten years had passed and the firms market share totally declined. For increase of the
market share the company thought they should concentrate more on the image of the
company from the scratch.
Questions
Q1. How will you help the firm to achieve its objective? Is a study required?
Q2. In your opinion what type of study is required? If possible, outline a research proposal
for the client.
2.10 SELF-ASSESSMENT QUESTIONS ________________________________
Q1. 'Market research is the corner stone of marketing.' Comment
Q2. Market research helps in 'decision making'. Elaborate
Q3. Client - researcher interface is a must for research? If yes, explain in detail.
Q4. Market research has its application areas and limitations. Comment
Q5. What is meant by research management?
36
Marketing Research
3.1 INTRODUCTION
When the researcher has analysed the problem and developed a research design
including the questionnaire, he has to decide whether the information is to be collected
from all the people comprising the population. In case the data to be collected from each
member of the population of interest, it is known as the census survey.
If the data is to be collected only from some members of the population, it is known as the
sample survey. Thus, the researcher has to decide whether he wants to conduct a census
or a sample survey to collect the data needed for his study.
In this unit, we will discuss some basic aspects of sampling. As we are concerned with the
practical aspects of sampling rather than the theoretical considerations, it would be
interesting to know why we should use sampling.
3.2 WHAT IS SAMPLING? __________________________________________
Sampling is a familiar idea with all of us as it frequently occurs in the course of our daily
events. Judgments are arrived at the entire number of items (population) on the basis of
analysis of a limited number of items (sample) e.g. a housewife tasting a spoonful of curry
to check the flavour and readiness of the vegetables, a fruit-seller slicing a fruit to show the
ripeness of the fruit to his customers, an executive doing a test drive of a car before
purchasing it. These are examples of sampling for the personal decision-making done in a
haphazard manner and with substantial risk of the sampling error.
However, research requires sampling to be done in a scientific manner for the validity and
reliability of results that have far reaching effects.
3.3 REASONS FOR SAMPLING _____________________________________
1. Decision makers have a limited time frame to take a decision. It is impossible to

collect information on the entire population during such a period, hence sampling is
used.
2. The budgetary constraints may prohibit the entire population study.
3. If the universe/population size is large, greater is the chance of using sampling.
4. Sampling is used when measuring a particular element of the product would render
them useless after examination. For example, testing of a photographic film, the useful
life of a fuse, etc.
40
Unit 3 Sampling
5. When an approximate idea serves the purpose rather than exact numbers.
6. The impossibility of finding all the population whereof e.g. plants.
3.4 ESTIMATING AND TESTING OF HYPOTHESIS

This stage involves the distinguishing of two objectives of sample surveys:
a) Population parameters
b) Testing a hypothesis
Estimation of a parameter refers to a situation in which the presence of a given population

is to be estimated. For example, one may be interested in the annual expenditure incurred
on tea drinkers or the proportion of workers working overtime in an industrial unit. In the
first example, the parameter refers to the average tea drinkers and in the second example,
the proportion of employees working overtime. Now to estimate a parameter, first a sample
is chosen, the elements in the sample are contacted and the necessary information is collected
from them.
From the information gathered, the statistics (average or proportion) is calculated. This
statistic is used as an estimate of the population parameter.
The second objective of sample surveys may be to test a hypothesis involving a comparison
of two or more numerical values. For example, if one would like to test the hypothesis that
at least 60 % of households have an internet connection in a town.
A sample survey is undertaken and the relevant survey data reveals that the overall percentage
is 55. The question now is whether these two percentages are significantly different.
3.5 DEFINITION AND MANAGERIAL OBJECTIVE OF SAMPLING

Sampling should provide the following benchmarks:
1. The data is representative of the population concerned.
2. Sufficient occurrence in a sample should exist to provide consistent results.
3. Using research resources as efficiently as time requirements permit.
41
Marketing Research
Definition of Sampling terms ••-*>•>.••:. >-<
Element: Element the unit about which information is collected for analysis.
Population/Universe: Population / Universe is an aggregation of elements over a period

of time, space extent and sampling unit. For example, female of 18 years (elements) living
in hostels (unit) in Pune (extent) in the year 2003 (time).
Sampling Unit: Elements) available for selection in some stage of the sampling process.
Survey Population: Is the aggregation of elements from which the sample is actually
chosen.
Census : Census is the complete enumeration of population or groups at a point of time

with respect to well-defined characteristics such as population, production, traffic on
particular roads. In some connection, the term is associated with the data collected rather
than the extent of the collection so that the term 'Sample Census' has a distinct meaning.
The partial enumeration resulting from a failure to cover the whole population, as distinct
from a designed sample enquiry, may be referred to as an incomplete census.
Sample: A part of a population, or a subset from a set of units, which is provided by

some process or other, usually by a deliberate selection with the object of investigating the
properties of the parent population or set.
Sample Survey: A survey which is carried out using a sampling method, i.e. in which a
portion only, and not the whole population, is surveyed.
Frame : A list, map or other specification of the units which constitute the available
information relating to the population designated for a particular sampling scheme. There
is a frame corresponding to each state of sampling in a multi-stage sampling scheme. The
frame may or may not contain information about the size or other supplementary information
of the units, but it should have enough details so that a unit, if included in the sample, may
be located and taken up for inquiry. The nature of the frame exerts a considerable influence
over the structure of a sample survey. It is rarely perfect, and may be inaccurate, incomplete,
inadequately described, out of date or subject to some degree of duplication. Reasonable
reliability in the frame is a desirable condition for the reliability of a sample survey based on
it
In multi-stage sampling, it is sometimes possible to construct the frame at higher stages

during the progress of the sample survey itself. For example, certain first stage units may
42
Unit 3 Sampling
be selected in the first instance; and then more detailed lists or maps be constructed by the
compilation of the available information or by direct observation only of the first-stage
units actually selected.
Sampling Error: Sampling error is that part of the difference between a population
value and an estimate thereof, derived from a random sample, which is due to the fact that
only a sample of values is observed; as distinct from errors due to imperfect selection, bias
in response or estimation, errors of observation and recording, etc. The totality of sampling
errors in all possible samples of the same size generates the sampling distribution of the
statistic which is being used to estimate the parent value.
Five steps in Sample Planning
1. Define the target population
2. Establish a "frame" of that population
3. Choose the method of selecting the sample unit
4. Determine the sample size
5. Selection of the actual (sample) members
1) Defining the Target Population: Sampling is intended to gain information about a

population. So it is critical at the outset to identify the population clearly and accurately.
Often target population identification is ignored as it appears obvious and
non-controversial, but attention to selection pays off.
Guidelines for identification of the target population
1. Well-defined research objectives lead to clear target population definition as

well.
2. Consider alternative, reasonable, target population definitions
3. If the research objectives is to learn the market response to the marketing

programme then detailed knowledge about the market is essential. Otherwise
the population definition will unnecessarily become broad.
4. Consider the appropriate sampling unit e.g. the individual / family as a unit.
43
Marketing Research
5. Specifications for exclusions in target population should be clear.
6. Over-definition of population makes the research costly and time consuming.

For example, all working wives between the age group of 25 - 40 years with an
income of Rs. 1 0,000 or more.
7. Consider the convenience: Preference should be given to a convenient sample if

it suits the research purpose. For example, a study on noise pollution.
Respondents will be best chosen from a township adjoining an industrial area.
2) Establish the framework of population : The sampling frame is usually a list of

population members used to obtain a sample. These boundaries could be geographic
or any distinguishable feature. For physical sciences, boundaries may not be significant
but for marketing they are a must.
3) Choose the method of selecting sampling unit : Sampling units may be selected
in a number of ways. It depends on the nature of the research, time and cost constraints,
research objectives and the expertise of the interviewer.
4) Determine the sample size : In order to make the survey, we have to determine
the sample size. For example, if one has to conduct the survey of teenagers smoking
cigarettes in a particular area, then how much sample size is to be considered for
giving the proper result, whether to choose 50 or 70 or 100 etc.
5) Selection of the actual sample member : As discussed above in point 4, out of

the 70 samples selected, which samples should be taken for hypothesis. The entire
sample cannot be taken and many of them may give the wrong interpretation.
Activity A ;
a) What do you understand by the

following?
• Population
• Elements
• Frame
44
Unit 3 Sampling
b) What are the various steps involved in the sampling process? Give any three important
steps.
Sampling Design
Sample Design
Non-Probability Samples Probability Samples
Convenience Simple Random
Snowball Quota Systematic Area
Judgment Stratified
1. Non-Probability samples: These are in which specific members are selected in

a non-random fashion as per convenience or due to a deliberate choice as per plan.
to a) Convenience sampling: The researcher chooses respondents on the basis

of convenience or accessibility. It is used for exploratory research.
b) Snowball sampling: In this method, the initial sample size is selected by using
probability methods but additional units are obtained through the information
supplied by initial units (referrals). This technique is used to locate sampling units
which have similar characteristics but are difficult to find. The referral process
may bring an element of bias in selection.
c) Judgement sampling: This method is of sampling by opinion. The researcher

who is acquainted with the population may decide arbitrarily, as per his judgement,
the units which represent the parameter of the study. Another form of judgemental
sampling is statistical judgement. In this method, the selection criterion is specified,
data on these factors are scrutinised and the elements meeting these specifications
are chosen for the same. The choice is (purposive) deliberate not random. This
method is valid when adequate data is available for the whole population
parameters and the individual elements in it.
45
Marketing Research
d) Quota control sampling: Is judgemental sampling with the constraint that the
sample includes minimum number from each specified subgroup in the population.
Demographic data such as geography, location, age, sex, education and income
group are used as qualifying parameters. The quota control has potential
weaknesses like
i Field workers may be unscrupulous in finding the exact sampling units

and numbers for the survey.
ii Population with listed characteristics may be hard to identify.
iii. Data available for classification may be obsolete and incorrect. For example,
population census.
Quota controls are used for national panel studies. A panel is a semi permanent
sample whose members may be included repetitively for successive studies.
2. Probability samples: In Probability samples, each population has a known

chance of being selected for inclusion in the study.
a) Simple random sampling is an approach in which each element has equal

probability of being selected. Such numbers may be selected through random
number tables or generated computer. For example, a lottery draw, dice.
b) Systematic sampling involves spreading the sample through the entire list of
population following a regular pattern.
e.g. Total population = 100 customers = Every 1

Random unit Otn'customer to get a prize 10
Total No. of winners
c) Stratified sampling : In stratified sampling, the population is divided into

subgroups (strata) within the sample framework and random samples are drawn
to improve the efficiency of the sampling plan and its reliability with a smaller
sample size. Sample sizes tend to be inversely proportional to the cost of the
sampling. Smaller the cost, larger will be the sample size.
d) Cluster /Area sampling: In cluster sampling the population is divided into

subgroups called clusters. A random sample of the subgroups is selected and all
tit ».
16-
Unit 3 Sampling
members of the subgroups become a part of the sample. This method is useful
when subgroups can be identified which are representative of the whole
population.
JS$ Activity B;
Give one example of each of the following sample designs:

• Cluster sampling
• Stratified sampling
• Simple random sampling
• Quota sampling
• Judgement sampling
3.6 PROBLEMS ASSOCIATED WITH SAMPLING ______________________
The accuracy and reliability of sampling data are affected by two different types of errors,
i.e. sampling error and non-sampling error.
Sampling Error: Sampling error occurs when the selected samples are not representative
of the population. Regardless of the care taken in sampling, some amount of sampling
error exists. It is the result of chance and subject to the law of probability. A sampling error
may be reduced by increasing the size of the sample. This is applicable to probability
sampling only.
Non-Sampling Errors: The name suggests all kind of errors except sampling errors.
These errorsjnclude,
respondents due to poor recall, inaccurate memory, bias,
2. Deceit by respondents when they do not want to

admit lack of knowledge, social status issues, taboos, etc.
47
Marketing Research
3. Poor sampling design, e.g. inaccessible respondents
due to wrong addresses, instructions, etc. "
4. Unfinished interviews due to
length/boringness/tocipersonal nature of the interview.
5. Recording errors, coding and editir
6. Misinterpretation due to ambiguous wording or lack of proper instructions.
Lipstein offers rules for minimising non-sampling errors in a survey. They are as
follows:
1. Sample survey should be as easy to execute as possible.
2. Use the smallest sample consistent with the research objectives.
3. Restrict the questionnaire to the data essential to the study.
4. Pre-test the questionnaire to check respondent fatigue.
5. Keep the interviewer and the respondent involved.
6. Don't ask consumers the questions which they really cannot answer.
7. Don't ask the interviewer to do the impossible. It encourages cheating.
3.7 NON-RESPONSE ISSUES IN SAMPLING

Non-response can be broadly classified into two problem area
Low response rates

• Refusal
• Inaccessibility
Refusal can be minimised by ,
1. Prior notification on the imminent survey rather than taking the respondent by surprise.
2. Motivating the respondents to participate in the survey by increasing their interest

and involvement.
3. Monetary and non-monetary incentives may be provided. For example, coupons,

pens, books, etc.
48
Umt 3 Sampling
4. Questionnaire design and administration should be brief, clear, low on fatigue quotient
and should be able to cover sensitive issues in a non-offending manner.
5. Follow up / reminder should be given to respondents after initial contact to submit the
completed forms.
6. Personalisation of letters increases the response rate.
In accessibility / not at home phenomenon
1. Prior appointments should be made.
2. Periodic follow up/call back should be attempted to get a response from the respondent.
Determining the sample size-adhoc methods
1. Rules of Thumb: The researcher suggests that the sample should be large enough
so that when it is divided into groups, each group will have a minimum sample size of
100 or more.
2. Budget constraint: The sample size will be determined by the total expenditure
one can incur for the process. However, it cannot guarantee the adequacy of the
sample size.
Total money to be spent
N = Cost / unit coverage
3. Comparable studies : Another approach is to find similar studies and use their
sample sizes as a guide.
3.8 FACTORS DETERMINING THE SAMPLE SIZE ___________________

1. Number of groups and subgroups that will be analysed.
2. Value of information and study in general and the level of accuracy required.
3. Cost of the sample. A cost benefit analysis must be done.
4. Variability of the population. Greater the variance, larger will be the sample size
required.
49
Unit 3 \ Sampling
• Since the magnitude of operations involved in a sample survey is small, both the
execution of the field work and the analysis of the results can be carried out speedily.
j • Sampling results in a greater economy of effort as a relatively small staff is

required to
I carry out the survey and to tabulate and process the survey data.
>
• f • A sample survey enables the researcher to collect more detailed information than
would otherwise be possible in a census survey. Also, information of a more specialised
type can be collected, which would not be possible in a census survey on account of
the availability of a small number of specialists.
• Since the scale of operations involved in a sample survey is small, the quality of the
interviewing, supervision and other related activities can be better than the quality in
a census survey.
Limitations of Sampling
a) When the information is needed on every unit in the population such as individuals,
dwelling units or business establishments, a sample survey cannot be of much help for
it fails to provide information on individual count.
b) Sampling gives rise to certain errors. If these errors are too large, the results of the
sample survey will be of extremely limited use.
c) In a census survey it may be easy to check the omissions of certain units in view of
complete coverage, this is not so in the case of a sample survey.
3.10 THE SAMPLING PROCESS _________________________

In this unit, we have learnt about the major advantages and limitations of sampling. We
now turn to the sampling process. It is the procedure required right from defining a population
to the actual selection of sample elements. Following are the steps involved in this process.
1. Define population : It is the aggregate of all the elements defined prior to the
.1$ selection of the sample. It is necessary to define population in terms of
• Elements
• Sampling units
• Area
• Tirneframe
51
0{iit 3 Sampling
• Since the magnitude of operations involved in a sample survey is small, both the
execution of the field work and the analysis of the results can be carried out speedily.
• Sampling results in a greater economy of effort as a relatively small staff is required to

carry out the survey and to tabulate and process the survey data.
• A sample survey enables the researcher to collect more detailed information than
would otherwise be possible in a census survey. Also, information of a more specialised
type can be collected, which would not be possible in a census survey on account of
the availability of a small number of specialists.
• Since the scale of operations involved in a sample survey is small, the quality of the
interviewing, supervision and other related activities can be better than the quality in
a census survey.
Limitations of Sampling
a) When the information is needed on every unit in the population such as individuals,
dwelling units or business establishments, a sample survey cannot be of much help for
it fails to provide information on individual count.
b) Sampling gives rise to certain errors. If these errors are too large, the results of the
sample survey will be of extremely limited use.
c) In a census survey it may be easy to check the omissions of certain units in view of
complete coverage, this is not so in the case of a sample survey.
3.10 THE SAMPLING PROCESS _____________________________ _ ^

In this unit, we have learnt about the major advantages and limitations of sampling. We
now turn to the sampling process. It is the procedure required right from defining a population
to the actual selection of sample elements. Following are the steps involved in this process.
1. Define population : It is the aggregate of all the elements defined prior to

the selection of the sample. It is necessary to define population in terms of
• Elements
• Sampling units
• Area
• Timeframe
51
Marketing Research
Let us consider an example. If you were to conduct a survey on the consumption of

Taj Mahal tea in the state of Maharashtra, then the specifications might be as follows:
• Element: Housewives
• Sampling units: Households/housewives
• Area: State of Maharashtra
• Time frame: April 1 to 15,2006
If a launching of a new consumer appliance/product is done, then the population will

change. It will now become
» Element: Product (Washing Machine)

• Sampling units: Retail outlets, super markets
• Area: Mumbai and the suburbs
• Time frame: April 7- 30,2006 . , I
It may be emphasised that all these four specifications must be contained in the
m designated population. Omission of any of them would render the definition of
population incomplete.
2. Identify the Sampling frame : Which could be a telephone directory, a list of

blocks and localities of a city, a map or any other list consisting of all the sampling
units. It may be pointed out that if the frame is incomplete or otherwise defective,
sampling will not be able to overcome these shortcomings.
The question is - how to ensure that the frame is perfect and free from defect. A
perfect frame is one where every element appears on the list separately, once, only
once, and nothing else appears on the list. This type of perfect frame would indicate
one-to-one correspondence between frame units and sampling units. But such perfect
frames are rather rare. Accordingly, one has to use frames with one deficiency or
another, but one should ensure that the frame is not too deficient so as to be given up
altogether.
This leads to several questions
• What are the criteria for a suitable frame?
• Does it adequately cover the population to be surveyed?
52
Unit 3 Sampling
• How complete is the frame?
• Is every unit that should be included represented?
• Is it accurate?
• Is the information about each individual unit correct?
• Does the frame as a whole contain units which no longer exist?
• Is there any duplication? If so, then the probability of selection is disturbed as a

unit can enter the sample more than once.
• Is the frame up-to-date? It could have met all the criteria when compiled but
could well be deficient when it came to be used. This could well be true of all
frames involving the human population as a change is taking place continuously.
• How convenient is it to use? Is it readily accessible?
• Is it arranged in a way suitable for sampling?
• Can it easily be re-arranged so as to enable us to introduce stratification and to

undertake multi-stage sampling?
These are demanding criteria and it is most unlikely that any frame would meet them
all. Nevertheless, they are the factors to be borne in mind whenever we undertake
random sampling.
In market research, most of the frames are from census reports, electoral registers,
lists of member units of trade and industry associations, lists of members of professional
bodies, lists of dwelling units maintained by local bodies, returns from an earlier survey
and large scale maps.
3. The Sampling unit: The sampling unit is the basic unit containing the elements of
target population. The sampling unit may be different from the element. For example,
if one wanted a sample of housewives, it might be possible to directly have access to
such a sample. However, it might be easier to select households as the sampling unit
and then interview housewives in each of the selected households.
The sampling frame should be complete and accurate, otherwise the selection of the
sampling unit might be defective. It is necessary to get a further specification of the
53
Marketing Research
sampling unit, both, in personal interviews and in telephonic interviews. Thus, in personal
interviews, a pertinent question is—of the several persons in a household, who should
be interviewed? If interviews are held during office timings when the heads of families
and other employed persons are away, interviewing would under-represent the
employed persons and over-represent elderly persons, housewives and the
unemployed. In view of these considerations, it is necessary to have a random process
of selection of the adult residents of each household. One method that could be used
for this purpose is to list all the eligible persons living at a particular address and then
select one of them.
4. The Sampling method: It indicates how the sample units are selected. One of the
most important decisions in this regard is to determine which of the two—probability
and non-probability sample is to be chosen. Probability samples are also known as
random samples and non-probability samples are known as non-random samples.
In case of a probability sample, the probability or chance of every unit in the population
being included in the sample is known. Further, the selection of specific units in the
sample depends entirely on chance. No substitution of one unit for another is
permissible. This means that no human judgement is involved in the selection of a
sample. In contrast, in a non-probability sample, the probability of inclusion of any
unit in the population in the sample is not known. In addition, the selection of units
within a sample involves human judgement rather than pure chance.
In case of a probability sample, it is possible to measure the sampling error and

thereby determine the degree of precision in the estimates with the help of the theory
of probability. This theory also enables us to consider, from amongst the various
possible sample designs, the one that will give the maximum information per rupee.
This is not possible when a non-probability sample is used.
Probability sampling enables us to choose representative sample designs. It also enables

us to estimate the extent to which the results based on such a sample are likely to be
different from what we would have obtained, had we covered the population in our
study. Conversely, the use of probabiliry sampling enables us to determine the sample
size for a given degree of precision, indicating that our sample results do not differ by
more than a specified amount from those yielded by a study covering the entire
population.
Although non-probability sampling does not yield these benefits, on account of its
convenience and economy, it is often preferred to probability sampling. If the researcher
is convinced that the risks involved in the use of a non-probability sample are more
54
Unit 3 Sampling
A compared to the offset being relatively cheap and convenient, his choice should be in
favour of non-probability sampling.
There are various types of sample designs which can be covered under the two
broad groups, which are,
a) Random or probability samples
b) Non-random or non-probability samples
5. Determine the sample size: In other words, one has to decide how many elements
of the target population are to be chosen.
6. The Sampling plan: This means that one should indicate how decisions made so
far are to be implemented. For example, if a survey of households is to be conducted,
a sampling plan should define a household, contain instructions to the interviewer as
to how he should take a systematic sample of households, advise him on what he
should do when no one is available on his visit to the household, and so on. These are
some pertinent issues in a sampling survey to which a sampling plan should provide
answers.
.*
7. Selection of the sample: This is the final step in the sampling process. A good deal
of office and fieldwork is involved in the actual selection of the sampling elements.
Most of the problems in this stage are faced by the interviewer while contacting the
sample-respondents.
JS$ Activity D;
You have been assigned to do a study on a brand which is fast losing its high profile image.
How will you design the sampling design in order to get the desired results to be achieved
for your client? At least design four major ones to which you will give prime importance.
Characteristics of a Good Sample Design
A good sample design requires the balancing of four broad criteria:
55
Marketing Research
• Goal orientation
• Measurability
• Practicality
• Economy
Goal orientation
Goal orientation suggests that a sample design should be oriented to the research objectives,
tailored to the survey design, and fitted to the survey conditions. If this is done, it should
influence the choice of the population, the measurement, and also the procedure of choosing
a sample.
Measurability
A sample design should enable the computation of valid estimates of its sampling variability.
Normally, this variability is expressed in the form of standard errors in surveys. However,
this is possible only in the case of probability sampling. In non-probability samples, such
as a quota sample, it is not possible to know the degree of precision of the survey results.
Practicality
This implies that the sample design can be followed properly in the survey, as envisaged
earlier. It is necessary that complete, correct, practical and clear instructions should be
given to the interviewer so that no mistakes are made in the selection of sampling units and
that the final selection in the field is not different from the original sample design. Practicality
also refers to the simplicity of the design, i.e. it should be capable of being understood and
followed in the actual operation of the field work.
Economy
Economy implies that the objectives of the survey should be achieved with minimum cost
and effort. Survey objectives are generally spelt out in terms of precision, i.e. the inverse
of the variance of survey estimates. For a given degree of precision, the sample design
should give the minimum cost. Alternatively, for a given per unit cost, the sample design
should achieve maximum precision with the minimum variance.
It may be pointed out that these four criteria come into conflict with each other in most of
the cases, and the researcher should carefully balance the conflicting criteria so that he is
able to select a really good sample design. As there is no unique method or procedure by
56
Unit 3 Sampling
which one can select a good sample, one has to compare several sample designs that can
be used in a survey. This means that one has to weigh the pros and cons, the strong and
weak points of various sample designs in respect of these four criteria, before selecting the
best possible one.
3.11 SUMMARY _______________________________________________

In this unit, we have seen how sampling plays an important role in a study, or rather, that
sampling designs play a crucial role in any study. One should be able to determine the
correct size of the sampling, because in the study if the sample answers are wrong, the
entire results are going to be affected. This results in the objective of the study not being in
the line of the objectives laid down.
3.12 KEY WORDS ______________________________________________

Hypothesis: proposition or supposition made as the basis for reasoning or investigation
Enumeration: specify or count
Probability: being probable. Likelihood of something happening
Orientation: person's attitude
3.13 CASE STUDY _____________________________________________

Hotel industry
A three-star hotel located in Mumbai has been experiencing a decline in its occupancy
during the past one year. The management has recently reviewed the problem and is
seriously considering attracting business executives and also providing adequate facilities
for holding business conferences, workshops, etc.
Since this would involve some renovation of the existing building in addition to new furniture
and equipment, the management wants to be cautious in undertaking such an expenditure.
Since its inception several years ago, the hotel has been maintaining a complete record of
its guests. When a person visits the hotel for the first time, details such as his name, age,
sex, permanent address, purpose of visit and duration of stay along with dates are entered
on a card. All subsequent visits along with the duration of stay are dated and recorded on
the same card.
57
Marketing Research
4.1 INTRODUCTION
A questionnaire is an instrument for seeking and recording data, either for interviewing or
for observation purposes as required for a meaningful measurement of data.
All three media of communication with respondents i.e. personal, telephone and mail, rely
on data collection forms known as questionnaire.
Definition
A questionnaire is simply a formalised schedule or a series of questions, written or verbal,

to obtain and record specified and relevant information from respondents with tolerable
accuracy and completeness.
4.2 FUNCTIONS AND CLASSIFICATION OF A QUESTIONNAIRE
1. It must translate information needs into a set of questions that the respondents can
and will answer.
2. It must motivate respondent to cooperate and complete the interview and trust that
confidentiality is ensured.
3. It must minimize response error. Response should be stimulated through greater
introspection, memory plumbing and record references.
4. To provide a system for recording, classifying and verifying data as per need.
5. To ensure consistency, if there are multiple interviewers.
6. Give instructions on what is wanted and the manner of responding.
7. It must facilitate data processing and analysis requirements.
Classification of Questionnaire
Degree of formality: Structured or Unstructured

Research Objective: Disguised or non disgui sed
Communication Method
Personal Mail Telephonic
Surveys interviews Diary
Panels
Unit 4 Questionnaire - Design Development
Interviewer
Administered Self Administered
A. Questionnaire design
In the research process, the design of the questionnaire is highly iterative, since it is an
integral part of research design. The objective is to seek consistency with research objectives,
budget, methods of data collection, data analysis and the respondents' willingness to answer
questions on the subject. Under these constraints, a researcher practices the art of
questionnaire writing for a good survey result. Success comes with experience and
empathising with the respondent and a clear understanding of the research objectives. The
process of designing has been shown as sequenced steps.
4.3 FLOW CHART FOR QUESTIONNAIRE DESIGN _______ .

• Determine the specific data to be sought
• Determine interviewing process
• Evaluate the question content
• Determine the response format
• Determine the wordings of the questions
• Determine the questionnaire structure
• Determine the physical characteristics of the form
• Eliminate bugs by pre-testing
• Revision
• Final draft
1. Determine the specific data to be sought
This step is self-evident and a prerequisite to designing effective questionnaire. One

must determine what exactly needs to be measured. Three key factors that need to
be defined are as follows:
• Statement of research purpose

• List of information to be measured (product, audience)
63
Marketing Research
• Draft of analysis plan (statistical methods to be used)
Omission of these steps can lead to inappropriate data collection and problems in the
analysis.
2. Determine the interviewing process

A problem confronting a researcher is in determining the type of questionnaire in
relation to the communication media and administration. They in turn depend on the
type of information sought and the respondent.
Personal Interview: 'Interviewer Administered' surveys much anticipate the

needs of the 'interviewer' and the 'interviewee' must. The colour code for the
interviewer and respondent instructions must be different so that the interviewer does
not read instructions meant for them. The question intent and wording should be
clear. Multiple-choice questions should be aided with visual aid for the respondent's
help. Flexibility should be adopted to meet specific needs of the respondent. Care
should be taken to prevent respondent fatigue and di si nterest.
'Self-Administered' personal surveys are those in which the respondent is given the
form by the interviewer to fill themselves. They may be sent through post also. They
are called mail surveys then. In such questionnaires, statement to invite co-operation
and instructions for filling the form must be written. Clarity of questions is essential.
The questionnaire must be visually attractive and easy from a respondent's point of
view. In case of a mail survey a prepaid envelope for reply increases the chance of
the return of form from the respondents.
Informal Interview is an unstructured format of data collection. It is also called an
in-depth interview. The interviewer provides the topic for discussion and encourages
the respondent to give complete information as per his/her knowledge. The interviewer
may stimulate the discussion/interview for a more orderly recording.
Telephone interviews, these interviews depend on the verbal clues only, hence
rating questions must be avoided. The questionnaire should be short, simple and not
time consuming. The pace of the conversation should be such that can hold the attention
of the respondent on the telephone.
3. Evaluate the question content
A clear statement of problem definition and objectives of study are necessary for the
guidance of the project and taking a decision of question content. A good questionnaire
can provide meaningful data and needs to be evaluated on three criterion.
64
• Does the respondent understand the question?
• Does he/she have necessary information to answer the question?
• Will the respondent provide the necessary information?
To decide the individual question content the researcher must ask himself:
• If the question is necessary?
• Are several questions needed instead of one to obtain the required information?
• Is the question ambiguous, double-barreled or confusing? Does it have multiple

answers?
e.g. In which departmental stores do you usually shop?
• Usually - The time frame of usual may be different for different individuals.
• Which shops are Departmental stores?
• Is one talking of stores in the city, in the locality or the entire universe?
Does the respondent have necessary information to answer the question?
Two reasons why respondents may give inadequate answers could be because of
lack of knowledge or lack of recall. The questions then need to be asked are as
follows:
• Is the respondent informed ? e.g. asking a housewife on insurance policy

premium. If the person designate in the household is asked this question then the
•' chances for accuracy of response are higher.
• Questions on issues on which no factual knowledge exists? e.g. will the prices of
gold rise? Only experts can give an estimated guesswork to such questions.
• What is the level of recall required to answer the question? e.g. what menu for
food did you order last month in this hotel?
• Can the respondent be articulate? Sometimes the respondents are not able to
express themselves, especially on open-ended questions.
65
Marketing Research
• The respondent may be unwilling to admit his/her lack of knowledge. Omissions,

telescoping and the creation of events may take place.
Will the respondent provide the information?
Sometimes even if the respondent can answer the questions accurately he/she refuses
to answer the questions :
• If it invades their privacy, e.g. questions on sex, income levels.
• Are socially sensitive issues, e.g. consumption of alcohol.
• If time and effort required to provide the information is too much. For example,
name of the product, brand, size, model number, cash/credit purchase, retailer's
name, etc.
• If the respondent sees no legitimate purpose. For example, a company wanting

information on income, occupation etc.
• Certain questions seem appropriate in certain contexts and not otherwise, for
example, a medical association asking about hygiene factors is acceptable but a Oi
fast food centre asking questions on hygiene may been inappropriate to a
respondent.
• Fears and suspicions produce non-response error.
• Personal prestige or social acceptability cause exaggeration errors. For example,

have you visited Shanwarwada?
For information on sensitive issues
• Third person technique may be used.
• Response categories can be provided instead of specific figures. For example,

income Rs.10000/-
• Such issues can be placed at the end of the questionnaire.
• The question can be hidden among topics the respondent is willing to answer.
• The question can be prefaced with a statement of general interest.
66
4. Decide on the Question Response Format
Once the issue of content has been addressed the next step is to decide on the type of
questions/response to use. Questions can be classified into two groups:-
4.4 TYPES OF QUESTIONS

Open Ended Close Ended
1 Free response Dichotomous
2 Probing Ranking
3 Projective techniques Checklist
3(a) Association techniques Multiple choice
3(b) Construction techniques Scales
3(c) Completion techniques
Open ended
In this type of question the response is unstructured and spontaneous without a prompting
aid. It reflects that responses exist in a conscious state of mind. There may be three subtypes
of open-ended questions.
• Free response
• Probing
• Projective techniques
1. Free response
For example, what has been your shopping experience at 'Shopper's Stop' ?
(Free response)
What has been your experience of the sales promotion schemes at' Shopper' s Stop' ?
(Free response directed to one aspect)
What has been your shopping experience at 'Shopper's Stop'?
(Answer by respondent)
67
Marketing Research Un
2. Probing
In the above questionnaire, can we ask anything else? (Probing)
Advantages
• Open-ended questions help obtain unanticipated, real views of the respondent.
• They help in exploratory research and opening up during initial questions.

• They help develop a rapport with the respondent and get real 'consumer words'.
Disadvantages
• Create semantic problems.
• They are difficult to record, tabulate and analyse.
• Create potential interviewer bias.

• Higher weightage is given to people who are more educated, exposed and
articulate.
• Are a time consuming and costly exercise.
3. Projective technique
Collecting data through this approach consists of presenting relatively ambiguous

stimulus such as words, pictures, etc. and asking the respondent to describe or verbalise
them by what enters into their mind while they think about the object. The responses
are then analysed and interpreted according to a predetermined psychological
framework. It is presumed that respondents display true feelings, motives and
personality. Further, these can be sub grouped as (a) Association (b) Construction
and (c) Completion techniques.
a) Association technique
Respondent is shown a series of words and is asked to respond to each with the
first word that strikes his mind.
For example, words 'Black and Shiny,' Respondent: Sunsilk
Trade names, brand names, promotional slogans, etc. have been studied with
this technique.
68
b) Construction technique
Respondents are required to view an ambiguous stimulus situation and are asked
to respond to it. For example, a picture showing a woman shopping for cooking
oil at a general store. The respondents are then asked to tell the story from that
picture. The researcher draws conclusions about purchasing habits and
preferences from the response.
c) Completion technique
Respondents are provided with incomplete sentences which they can complete
as they choose to do.
• Groundnut oil is...
• A cooking oil should have...
In all these techniques the advantages and disadvantages are as follows:
Advantages
• Enables information on subjects not approachable through questioning.
• Provides insights for hypothesis building at the exploratory stage of research.
Disadvantages
• Experienced and highly trained interviewers are required to conduct such
techniques.
• Interpretation of results is highly subjective.
• Sample population may not be representative of the sampling universe.
Close ended
The question and responses are structured and closed. There are various types of
close-ended questions.
1. Dichotomous 2. Ranking
3. Checklist 4. Multiple choices
5. Scales
1. Dichotomous question
This type allows response in affirmative or negative. The situation of 'don't know'
cannot be recorded.
69
Marketing Research
1. Do you like this book? Ans. Yes No

Such questions may be used for qualifying a person to answer further questions.
For example, do you like this book Yes No
Go to question 2 Go to question 5
Advantages
t Easy to administer, tabulate, edit and analyse.
• Provide a lead to questions that may require greater details.
• Easy to answer.
» LOW interviewer bias due to specific response.
Disadvantages
9 Detailed information not possible.
• Forces respondent to make a choice though they may not think it an appropriate
response.
• Difficulty in wording properly which leads to 'first phrase', or 'central phrase'
bias.
For example, do you use Lifebuoy liquid or Lifebuoy soap?
2. Ranking Questions
Involves a respondent ranking comparatively the items listed. Several alternatives

have to be compared with each other at the same time.
For example, please rank the most important feature as number 1 and that of least
importance to you as number 4.
I prefer shopping at 'Shopping Glory' because it has:

Low prices _______________
Free home delivery _^______________
Convenient location _______________
Range of products _______________
70
Advantages
• Relatively easy to ask and tabulate.
• Respondents are generally familiar with this method.
• Response can be measured quickly.
Disadvantages .
• The number of items that can be ranked are limited (5-6 items). *
• It assumes that the respondent has complete knowledge about each item.
• It does not indicate the degree of preference intervals between ranks. '
3. Checklist * '
In a checklist form, a person checks one or more of responses that are listed for
answers. They are useful for factual data, demographic classification, etc. Care should
be taken to list major categories as respondents fail to write their response for others.
For example, in your home which of the domestic appliances are used? (Check as
many as applicable)
Microwave oven
Automatic washing machine '
Food processor
Telephone answering machine

Dishwasher
4. Multiple-choice answers
These questions list more than two answers for the choice of response. The respondent
selects the response that approximates best to his/her own view or can opt for don't
know/ can't say/ others. Please specify option. For example, which are the benefits
ofAmulMilk? <-<
1. Homogenised
Marketing Research
2. Pasteurised
3. Hygienically packed
4. Low price
5. Balanced fat
6. None of the above
7. Others
Advantages
1. They are quick and easy to ask
•. . . . - ,• - ' !
2. Editing, tabulation and analysis is faster.
3. Low on interviewer bias.
4. Overcomes problems of open-ended questions.
Disadvantages
1. Listing of all possible mutually exclusive alternatives is difficult.
2. Order of responses may create position bias.
3. Alternate responses may hold a different meaning for the
respondent.
Activity A;
Tick the odd one out

a. Free response
b. Probing
c. Projective techniques
d. Dichotomous
4.5 CHOICE OF SURVEY METHODS

As there are four methods by which data can be collected in a survey, it may be worthwhile
to know their relative strengths and weaknesses. The methods in question are - personal
survey, mail survey, telephone survey and computer survey. However, as computers in
interviewing respondents are not commonly used as yet in our country, the comparison
will be confined to the three methods, giving a comparative idea of three survey methods
on selected criteria.
Table 4.1: Comparison of survey methods

Criteria Personal Mail Telephone
Survey Survey Survey
Response rate Fair Poor Good
Time Required Slow Slow Fast
Cost High Low Low
Sample Control Good Poor Excellent
Supervision of Not
Field work Fair applicable Excellent
Quantity of info Good Limited Limited
Quality of info Good Fair Excellent
Versatility Excellent Fair Good
It is evident from the table above that none of these methods is perfect and free from all the
limitations. A method may be good or excellent in case of a few criteria, but in respect of
other criteria, its rating may be low.
The market researcher has to exercise great care in choosing the method on the basis of a
couple of criteria or considerations which he thinks are important in the survey. For example,
when a large quantity of information needs to be collected through a survey, telephone and
mail survey methods will be inappropriate. The choice should obviously be in favour of
personal survey. In contrast, if a short survey is to be conducted in a limited time, telephone
survey should have the preference. There is, of course, an inherent limitation of telephone
survey in our country, as telephones are not yet very common in cities, not to mention of
rural areas and the most importantly people do not know how to use the telephones.
73
Marketing Research
A. How to use the correct wording in the question?
The questionnaire is an interface between the researcher and the respondent. The respondent
will either read or hear the questionnaire. Improper wording can be a barrier in meaningful
data collection. Some principles to choosing the appropriate wording are:
1. Use simple language and ordinary words to meet the vocabulary level of the
respondent. For example, say fever instead of Pyrexia.
2. Avoid technical jargons and abbreviated forms. For example, LPG,MNC
3. Avoid ambiguous words like usually, normally, frequently, often, regularly,

sometimes, occasionally, etc.
4. Be as specific as possible.
For example, If somebody asks what kind of car do you own? The answer could be
on brand, model, body or type.
5. Avoid double barreled questions: a question that calls for several responses at once
is called a double barreled question. These type of questions should be avoided. For
example, do you use Dettol liquid and Dettol plaster?
6. Avoid questions for which the respondent has to make generalisations or guess
estimates. For example, TV programmes have you watched for the past year?
7. Avoid leading questions, which give a clue and bias the respondent on what the
answer should be. For example, do you think that the company has a poor after-
sales service record?
8. Avoid implicit assumptions.
9. Use positive and negative sentences to balance the image.
10. Avoid loaded questions. A loaded question has more subtle bias than a leading
question. Loading could result because of an emotionally toned word for example,
'oilmonopoly', 'luxury items' etc.
11. Avoid implicit alternatives that cannot be explicitly expressed in options.
12. Avoid lengthy questions.

Unit 4
Development
Questionnaire - Design
11
B. How to determine the questionnaire structure? » ,
Once a set of questions has been decided upon, they need to be put in a form that
facilitates administration - which questions should come first and which ones should
fall later. Thus, arises the issue of sequencing in terms of breadth/generality. The
general questions come first and the more specific questions for the later part of the
interview. This is called the funnel approach or the flowerpot sequences.
1. The opening questions must be simple, interesting and non-threatening so that

the rapport is built with the respondent. For example, dichotomous questions.
2. Qualifying questions serve as opening questions to find out whether the person
is the right type to provide the information.
3. Questions should be asked in a logical sequence.
4. Warm up questions direct thinking and memory. Thus the difficult, sensitive and
complex questions should be placed in the later part of the sequence.
5. General questions should precede specific questions.
6. Branching questions must cover all possible contingencies.
4.6 CLOSE ENDED AND OPEN ENDED QUESTIONS _________________
Questions can be classified in various ways. One way of classification is as follows:
• Open-ended questions
• Close-ended questions
An open-ended or simply 'open' or 'free answer' question gives the respondent complete
freedom to decide the form, length and detail of the answer. Open questions are preferred
when the researcher is interested in knowing what is uppermost in the mind of the
respondent. However, open questions pose certain problems. At the time of the actual
interview, it becomes difficult for the interviewer to note down the respondent's answer
verbatim. If the interviewer has to take down the answer all by himself without any
mechanical aid, he is quite likely to miss some vital information contained in the respondent's
answer. Further, if several interviewers are conducting interviews and each one is recording
the answers to opinion questions according to his understanding, and in his own way, then
there is likely to be an element of bias in the recorded answers. Another difficulty with
75
Marketing Research
respect to open questions is that it is extremely difficult to compress lengthy answers in a

meaningful manner. Such answers may be good qualitatively but their quantification becomes
extremely difficult, if not impossible. The dichotomous question has only two answers in
the form 'yes' or 'no', 'true' or 'false', 'use' or 'do not use', etc. An example of a
dichotomous question is:
Do you use Colgate in any way?
Yes__________ No !
There cannot be a third answer. However, in some cases, there may be a third answer
which may come from those who do not want to take a definite stand one way or the
other. For example, take the following question:
Do you like to watch Television? ..•••-• r
Yes . ___ No __________ Neither like nor dislike _____________
The third alternative may be included so as to provide for those respondents who do not
have a positive preference or aversion to movies.
It may be pointed out that dichotomous questions are most convenient or least bothersome
to respondents, who have simply to indicate their choice from the two possible answers.
As such, these questions require the minimum possible time of the respondents. Also,
answers to such questions are easy to edit, tabulate and interpret.
In the case of multiple-choice questions, the respondent is offered two or more choices.
The market researcher exhausts all the possible choices and the respondent has to indicate
which one is applicable in his case. For example, the following is a multiple-choice question:
Which of the following brand/brands do you use for washing clothes?
Rin _____ , Ariel ______ , Surf ______ , Super 777 Bar ______ , Wheel
Any other (please specify)
Obviously, the respondent is likely to take more time to answer a multiple-choice question
as compared to a dichotomous one. Also, more time is required in the editing, tabulation
and interpretation of data.
J&> Activity B;
a) Which type of questions may be used as branching out options as to ascertain the
requirement of the respondent in the sampling plan?
b) Typography is important in which step of the questionnaire design. ________
c) Attitudes are measured through ______________ and ______ techniques.
d) The questionnaire structure should follow the. . approach.
e) Third person techniques are used for getting response _____ issues.
on.
4.7 SAMPLE QUESTIONNAIRE

RURAL DENTAL CARE JN/60544
Extended Interview Questionnaire
Name of respondent.
Address ________
PUNE NASIK
Centers: Baramati 1 Junnar 2 3 4
Villages: Ambi 1 Udapur 6 Surgana 16
Malwadi 2 Ane 7 11 Umbarthan 17 18 19
Waki 3 Arvi 8 Bhormal 20
Kaatphal 4 12 Barhe
Yevla
Dorlewadi 5 13 Palsan
Manori
(Budrak) 14 Sarad
Nagde
Ajanwale 9 Ankai
Agar 10 Rajapur
Suthegaon 15
77
Marketing Research
Name of interviewed _Sign_ Date
Name of supervisory _Sign_ Date
Record responses in Marathi and Translate in English.
Space provided for both.
Classification Data: Approach Males / Females at Home
SHOW CARD A
Current product usage CURRENT IMPLEMENT USAGE
Stick 1 2 3 4 Finger 8
Salt 5 6 7 Brush 9
Home made powder
Red powder
Mashiri
White powder
Toothpaste INTERVIEWER TO
COLLECT PRODUCT Prod. Impl.
SPECIMEN 7-8
9-10
ii. SHOW CARD B
Age: Below 12 years 1 TERMINATE

12-17 years 2
18-25 years 3
26-35 years 4
36-45 years 5
46 + years 6
CONTINUE IF ANY OTHER THAN 1 CODED 1

1
Hi. SHOW CARD

C Illiterate
Education
1
Literate but no formal
78
education 2
School upto 4 yrs 3
5-9 yrs 4
SSC/HSC 5
Some college but not graduate 6
Graduate/PG diploma 7
Postgraduate 8 12
iv. SHOW CARD

D
if
I Marital Status: Single 1

Married 2
13
Divorced 3
Widowed 4
v. SHOW CARD E
URBAN INTERACTION
Visit town/city
Frequency City
3-4 times in a year 5 - 6 Taluka Pune/Nasik
times in a year 7-12 times 1 1
in a year Twice in a year 2 , ' 2
More than twice in a year 3 3
4 4
vi. Used most often at 5 5
14 15
home
Detergent/Washing powder 1 BRAND 16 -17
Toilet soap Shampoo
2 BRAND 18 - 19
Cooking oil
3 BRAND 20-21
4 BRAND 22 - 23
7
9
Marketing Research
vii. I buy these products mostly from:

Village shop 1
Nearest village 234
Go to town/city/taluka
Relatives/friends get it for me 24
Sex: Male: 1 Female: 2
25
viii. Media exposure
TV viewership Readership-Newspaper Radio-Listenership
Everyday Once in 1 Everyd Everyday Once in

2-3 days Once in ay 1 2-3 days Once in a
a week Once in 2 Once week
15 days Once in a in 2-3 4 Once in 15 days

month days Once in a month 2
3 Once Less than once a month 6
in a week 3 DK/DS
4 Once in 15 days
5 Once in a month 5 35
t
Less than once a month 6 Less than once a month 6
Never 7 DK/CS 7
DK/CS 8 Never 8
33 34
Main Interview
Namaste! I am from the Indian Market Research Bureau. We are currently conducting a
survey to understand people's habits and practices in dental care. In this connection.
I would like to talk to you for sometime.
Area 1. Current Practices X Pathways of Change
Q. 1 What are the things that you are doing currently to look after your teeth and
maintain oral hygiene? What products and implements are you using?
m
Products used Implements used

Stick finger
1 brush
A B C
Nothing
2 A B C
Kulla
Tobacco 3 A B C
Charcoal/ash 4 A B C
Salt 5 A B C
Home made vder A B C
powder irs6 A B C
Coloured 87 A B C
powders 9 A B C
Prod. Impl. - 3
White powders
Toothpaste 36 38 7
"1 -3
Combination: 1
Combination: 2
a. Please tell me why do you use these products/implements, for what sorts of benefits?
b. Do these products/implements create any problem?
c. What are the likely solutions to these problems?
Record Responses for Combinations Mentioned by Respondents
Combination: 1 ________________________________ _ ____________________

BENEFITS: ________________________________________________________
41-52
PROBLEMS:
53-60
SOLUTIONS :
81
Marketing Research
61-66
Combination: 2
BENEFITS: _
71-82
PROBLEMS:
83-90
SOLUTIONS:.
91-96
Q.2 ASK FOR EACH COMBINATION SEPARATELY

and. .MENTIONPRODUCT
You mentioned that you use.

AND IMPLEMENT FROM Q. 1. Could you tell me what influenced you to use this
product and implement? PROBE : Friend's/relative's advice; doctor's advice;
convenience; ads on TV/radio; visit to the city; price; product form (paste or powder).
Combination: 1. Combination: 2.
97-99 100 - 102
How much does. (PRODUCT) and. .(IMPLEMENT) cost?
Q. 3a
b How long does one pack
last? c From where do you
buy it?
Q. 3a Q- 3b Q. 3c Q. 3d
Cost Size/wt Duration Boughtfrom Started
(inRs.) (ingms) (indays) (Refercode using
list) at age (years)
111-125 Product 1:
126-140 Product 2:
141-155 Implement:
Toothbrush
Code List for Q. 3c
(Bought from)
Village shop 1
Talukashop 2
City shop 3
* Gifted 4
Q. 4a When you started using _____

PRODUCT and IMPLEMENTS
(USED CURRENTLY), what were your expectations from it?
AFTER SPONTANEOUS RESPONSE
PROBE: Fresh breath, clean teeth, no dental problems; whiteness; good taste;
convenience
!>
,!*iV
PRODUCT 1
SPONTANEOUS : __________________ . ................ .._- ..... , .........
PROBED: 156-161
162 -173
PRODUCT 2:
SPONTANEOUS :.
PROBED :_______ 174 - 179
180-191
Q. 4b Suppose the product had to be made even better, what would you suggest
to
improve it?
83
Marketing Research
PROBE^shbreafc clean .eete

convenience
192-203
Area II: Past Practices - Oral Care

Q.5 a. When did you first start to clean your teeth? Age specify.
b. What are all the things, that you have even done/used to maintain oral hygiene
and care of your teeth?
c. FOR ALL PRODUCTS USED ASK: What implements did you use to
apply these products?
d. At what age did you start using these products and implements?
Q.5(d)
Nothing Kulla
Tobacco
Charcoal/ash
Salt
Home made powders
Coloured powders
White powders
Toothpaste
Q.5(b) Q.5(c)
Products used Implements used
Age Stick Finger Brus
1 A B h
C
2 A B C
3 A B C
4 A B C
5 A B C
iers A B C
•s6 A B C
87 A B C
9 A B C
PI Age
209 -212
Q.6 You have mentioned that you used and and

MENTION PRODUCTS AND IMPLEMENTS
CODED IN Q.5(b) and Q.5(c). I would like to understand a little about why you
began to use these products and implements?
IF NO CHANGE CODED IN Q.5 (b) and Q.5(c), GO TO Q.7(c)
a. TAKE PRODUCTS AND IMPLEMENTS CODED FOR AGES

BELOW 15:
You used MENTION and PRODUCT

AND IMPLEMENT CODED, till you were 15
years old, how did you get to know about these things?
PROBE: Parents' gave it; doctor's advice; friend's advice; learnt about it in
school:
213-218
b. ASK Q.6(b), IF MORE THAN ONE RESPONSE OBTAINED IN Q
(a).
Which of these do you feel would be the most important reason for you to
have started using these products and implements?
219 - 224
c. TAKE NEXT SET OF PRODUCTS AND IMPLEMENTS CODED AND

ASK: INTERVIEWER TO ENSURE THAT IN NEW SET OF PRODUCT
AND IMPLEMENT AT LEAST ONE ELEMENT i.e., PRODUCT OR
IMPLEMENT SHOULD HAVE CHANGED. AT AGE ______ MENTION
AGE, you changed to a new product ______________ MENTION
PRODUCT and ____________ MENTION IMPLEMENT. Could you
tell me, why you made this change? What happened then? PROBE: relative
got it from the city; saw ads on TV and thought it was better; was available in
the shops; any other:
225 - 230 85
Marketing Research Unit 4
IF NO CHANGE IN PRODUCT OR IMPLEMENT GO TO Q.7a.

.(PRODUCT) and.
Q .7 a. You have been using _ .(IMPLEMENT)
for ___________ years now? Could you tell me, have you thought of
changing to any other product implement?
PROBE: Why? Why not?
231-236
b. You changed (PRODUCT) and IMPLEMENT to
________________________________________
. (CURRENT PRODUCT) and.
IMPLEMENT). Could you tell me why you changed to the product?
PROBE: Convenience; doctor's advice; availability; friend's advice; any

other?
Which of these would be the most important reason, according to you?
237-242
c. I have here on cards, reasons why people keep changing products and
implements over time to take care of their teeth. Can you tell me which of
these do you think influence this change?
SHOW CARD 1 to 14
Education 1 Flavour 89
Age 2 Modern 10
Urban influence 3 Scientific
Greater exposure to 11
mass media 4 Expensive 12 1314
Economy 5 Better quality
Product form i.e. paste/powder Fashion
6 Taste
Other (SPECIFY)
Dental problems 7
d. FOR ALL REASONS CODED IN Q.7(c) PROBE IN DETAIL

ASK: Could you tell me a little more about how you feel as to what has
changed dental habits?
MENTION REASONS CODED IN Q.7(c)
Education:
Age:
Urban influence: .
Greater exposure '

to mass media:
Economy:
Product form
i.e., paste/ powder:
Dental problems:
Ravour: Modern:
Scientific:
Expensive: Better
quality: Fashion:
Taste: Other
SPECIFY
87
Marketing Research
5.1 INTRODUCTION
In unit 1, we have seen the difference between primary and secondary data, its advantages
and disadvantages. We have also learnt the different observation methods. This unit will
help us to identify and select the primary and secondary data. We will also learn the
different scales to be used while designing the questionnaire and the usage of sampling
methods too.
Market research is often concerned with the behaviour of the consumer. In this respect,
market researchers have drawn heavily on the behavioural sciences such as psychology
and sociology. In fact, the contribution of these sciences to market research has been very
significant, especially with regard to these two aspects. First, the research techniques used
by psychologists and sociologists, being of a considerable relevance to market researchers,
have been amply used by them. Secondly, the concepts and theories of these behavioural
sciences have also proved to be relevant to market researchers.
If the attitude of the public towards a company or its product is unfavourable, the company
will not be able to sustain itself for too long. It is, therefore, in the interest of the company
to ensure that people have a favourable attitude toward its product. However, this by itself
is not enough. The company must also look into the future to anticipate the preferences of
the public. Thus, a study of attitudes becomes very relevant to market researchers. Another
area where attitudes seem to play a major part is advertising. In these days of increasing
competition, advertising has become almost unavoidable, and it makes use of various
psychological techniques. For example, take the case of copy research. A market researcher
would like to decide on the message the advertisement should carry, how it should be
conveyed and the most effective mode of taking this message to the consumer. To give yet
another example of the importance of attitude measurement in market research, take the
case of a manufacturer who wants to know how much his product is likely to sell. For this
purpose, he may have to undertake a research on buying behaviour. By observing customers
as they buy goods, he can learn their preferences for a particular brand. Apart from this
direct observation, which would need a long time, he may embark upon an attitudinal
survey. This would reveal the likes and dislikes of the customer in respect of a certain
brand, which would be extremely helpful to the manufacturer.
5.2 TYPES OF SCALES
The measurement of marketing phenomenon is fundamental to providing meaningful

information for decision making in marketing. Measurement transforms the characteristics
of an object into a form that can be analysed by research. For marketing use, they are
98
Unit 5 Scaling Methods
assigned numbers (quantitative data). However, the qualitative elements can be best
described as scales.
Scales are used to measure response. They are classified into four types:
1. Nominal
2. Ordinal
3. Interval
4. Ratio
1. Nominal scale
A nominal scale is one in which numbers are only used as labels and have no quantitative/
numerical significance. Variables such as sex, geographic location, occupation, religion,
brand awareness are studied under nominal scales. In these cases, mere counting or
numeration of individuals in each class is done. Statistics such as percentages, mode,
chi-square and binominal tests are possible. Nominal scales indicate that two/more
entities are different from each other.
For example, Sex

Male 1 OR Female 1
Female 2 Male 2
For example, Occupation

Service 1 Business 1
Business 2 OR Service 2
2. Ordinal scale
These help place different entities (objects, individuals or responses) in relative position
with respect to a particular aspect. These ranks are not interchangeable since they
possess a meaningful order. They indicate greater than and lesser than value between
variable(s). Though, by how much, is not yet known. Permissible statistics with ordinal
scales are percentiles, median and rank-order correlation. They are used to measure
attitudes, preferences, occupation, social class, etc.
For example, Bombay Dyeing asking respondents to show their preference of shirts
design for purchase.
99
Marketing Research
Design Ranking
Respondent Style 1 Style 2 Style 3
A 1 2 3
B 2 3 1
C 3 1 2
D 3 2 1
E 3 2 1
3. Interval scale
This scale is used to measure common statistical measures such as the arithmetic
mean, range, standard deviation and correlation. A researcher gauges preference.
liking or importance of a particular aspect of a brand on a continuous basis and
distance between one point on the scale to another (interval) is the same. It does not
help state that a brand is liked twice as much as some other brand since it does not
have the property of absolute magnitude i.e. zero point. These scales are used to
measure attitudes, opinions, index number, etc.
For example Respondent Willingness to buy

1 12345678910
2 12345678910
3 12345678910
4 123456789 10
We cannot say that the willingness of the two respondents with the score of 2 and 4
is same as the two respondents with the score of 8 and 10.
4. Ratio scale
Thise is the most powerful of the above scales. Unlike interval scale, it has a zero
point. It can be used for all kinds of statistical calculation. (Geometric mean, harmonic
mean, co-efficient variation). There is an equality of ratios. For example,
(1/2 = 2/4). They are used for variables such as age, number of customers, sales.
costs, height, length, etc.
Further, let us discuss in detail these four types of scales as explained with the help of
table 5.1 as shown below:
100
Table 5.1
Scale Characteristics Basic operation
Nominal No order, distance, or origin Determination of equality
Ordinal Order but no distance or Determination of greater or
Unique origin lesser values
Interval Both order and distance but no Determination of equality of

unique origin. intervals or differences
Ratio Order, distance, and unique origin Determination of equality or
ratios
Nominal Scales
Nominal scales are more widely used than any other scale for research in social sciences.
In such a scale, the numbers serve as labels to identify persons, objects or events. Thus,
numbers may be assigned to students in a class or patients in a hospital. We might further
use the nominal scale by counting students with a certain characteristic or attribute such as
those who reside in the university hostels, etc. In a nominal scale, we split a set into subsets
which are mutually exclusive and collectively exhaustive. Consider the following example:
Number of Students
Students Undergraduates Post graduates Total
Day Scholars 800 200 1000

Hostellers 400 150 550
Total 1200 350 1550
In this example, students have been identified and counted by two characteristics, namely,
whether they are pursuing an undergraduate or post-graduate course of study and their
place of residence. The numbers given in each of the four cells are mutually exclusive and
the total of these four cells gives the total number of students. In a nominal scale, the only
operation involved is the counting of numbers in each group. It is, therefore, the simplest of
the four scales and also the least powerful. The scale does not show any order or distance
relationship nor does it have any arithmetical origin. In view of these limitations of a nominal
scale, it is unsuitable in determining relationships but is very useful in preliminary or
101
Marketing Research
exploratory work, where it is sufficient to know the broad dimensions of a certain

phenomenon.
Ordinal Scales
Ordinal scales, as the name implies, are ranking scales. Besides having the unique
characteristics of the nominal scale, these scales also indicate the order. This is possible
when one is able to distinguish elements on the basis of a single direction. For example,
one may rank two or more households according to their annual income or expenditure.
Suppose we have five households with annual incomes as shown below:
Household A B C D E
Income (Rs) 6,000 4,800 12,000 11,000
10,000
If the household with the highest income is to be given No. 1 and the next to it as No. 2,
and so on, then the following order will emerge:
Household Order of households on the basis
of annual income
A D
B E
C B
D A
E C
This is the use of an ordinal scale, which involves the ordering of households on the basis
of their annual income. The point to be emphasised is that it is a mere ordering and does
not indicate the differences in annual income amongst the five households. On the basis of
this scale, one cannot say whether the difference in annual income between two households
D and E, is greater than, less than, or equal to the difference between any other two
households, say, E and B.
The use of the ordinal scale is possible when one is able to distinguish a certain product on
the basis of a particular attribute. The above example was numerical, dealing with the
annual income of households. It was simple as it did not involve any difficulty in ordering.
Consider another example where the acceptability of a soft drink is involved. Here, one
can ask the respondents, questions on the basis of one or more attributes such as flavour,
102
Jt.
colour, etc. Respondents may be asked to indicate whether they like the soft drink or not.
One can develop a five-point scale as given below:
I strongly like it +2
I like it somewhat +1
I am indifferent 0
I dislike somewhat -1
I strongly dislike -2
In this manner, ranking can be obtained by asking respondents their level of acceptability.
One can then combine the individual rankings and get a collective ranking of the group.
Interval Scales
The third type of scale is the interval scale. It possesses not only the power of the nominal
and ordinal scales but also additional strength, which is the determination of the equality of
differences. The classic example of an interval scale is the measurement of the temperature.
Both the Fahrenheit and Centigrade scales belong to this type. One can say, on the basis
of this scale, that a temperature of 100 degrees is 20 degrees warmer than 80 degrees and
20 degrees cooler than 120 degrees. It may be noted that differences between two values,
say, on a temperature scale, are multiples of each other. Thus, the difference between
40°F and 20°F is half the difference between 60°F and 20°F, using the conversion formula
from Fahrenheit to Centigrade.
Degrees centigrade = 5/9 (Degrees Fahrenheit - 32), we can find the

corresponding temperatures in Centigrade.
40° 5/9 (40 - 32)
40 = — =
4.44°C
5
20° /9 x 5/9( 2 0 - 3 2)
60
5/9
40°-20° (-12)=-— =-6.66°C
4.44 - (- 6.66)
4.44 +6.66 =11.10°C
= 5/9 (60 - 32) = 5/9 x 28 = 140/9 = 15.55°C
15.55-(-6.66) = 15.55+ 6.66
22.21 °C which is approximately twice of 11.10°C
Marketing Research
The above example shows that on a particular scale, equal differences indicate equal
differences in value with regard to that scale only.
Interval scales are more powerful than the nominal and ordinal scales. Also, they are
quicker to complete and researchers find them more convenient.
Ratio Scales
Ratio scales possess the powers of the preceding three scales as also the concept of
absolute zero or origin. Thus, they have order, distance and unique origin and are the most
superior amongst all the scales. Examples of ratio scales are the commonly used physical
dimensions such as height, weight, distance, money value and population counts. Equal
ratio on the ratio scale indicates the equal ratio among the elements being measured. For
example, 9 Ibs and 45 Ibs are in the ratio of 1:5. If we convert pounds into ounces, the
same ratio will be obtained. Thus, 144 ounces and 720 ounces have the same ratio of
1:5, as earlier. In other words, one can change over from one unit to another by using the
relevant conversion factor. In the above example, a change from pounds to ounces involved
the multiplication of the two figures by 16. This facility of conversion from one unit of
measurement to another is available in the case of the ratio scale alone.
From the foregoing account of the four types of scales, it should be clear that these scales
are in increasing order of sophistication from the viewpoint of data analysis.
Activity A;
Construct the Scales
a) Make a graphic scale to measure the attitude towards teenagers about eating in "Five
Star" hotels.
b) Construct an ordinal/ranking scale for the attributes of Maruti 800.
104
c) Chart out a semantic differential scale of a TV programme.
d) Illustrate an example using the likert scale.
5.3 THE CONCEPT OF ATTITUDE___________________________________

The meaning of 'attitude' is settled behaviour, as indicating opinion. However, this does
not help us much except that it implies in a given situation how a person will act naturally in
a certain manner, depending on his attitude.
In the late 20's the definition was defined as the overall behaviour of a man's inclinations
and feelings, prejudice or bias, preconceived notions, ideas, fears, threats and convictions
about any specific topic.
Attitude is a subjective and personal affair. The term 'opinion' symbolises an attitude. In
fact, it is the verbal expression of attitude.
Apart from this, the term 'attitude' has been defined in a different manner by different
authors.
Some say it is degree of positive or negative effect associated with the psychological
aspect.
Another author defines that it is a system which defines positive or negative evaluations,
emotions, feelings and the action tendency towards an object.
105
Marketing Research
5.4 CRITERIA FOR A GOOD TEST
The two important criteria for ascertaining whether the scale developed is good or not are
reliability and validity.
Reliability
By means of reliability we mean the measurements made under constant conditions giving
the exact results where no changes in the characteristics takes place.
This can be undertaken by using the same scale or test on the same set of respondents,
using the same methods. However, in practice, this becomes difficult especially as the
same set of respondents may not be inclined to answer the same questions again. Another
point to note is that the same respondents may remember their previous answers to the
questions and give the same answers.
This would indicate greater reliability of the scale or test than it may be in reality. The other
factor that might distort the criterion of reliability is that the respondents may become
extremely cautious and careful and may avoid giving accurate answers for the second time
or their attitude might have changed during the intervening period.
In either of these cases, the use of the reliability criterion will not be very helpful as the test
and retest scores will not be rigidly comparable. If the time period between the test and
retest is a short one, the respondents are likely to remember their earlier answers and as
such the memory effect may distort the reliability test though there would be meager chances
of change in the respondents' attitude. On the contrary, if there is a long interval between
the test and the retest, developments during this interval may lead to a change in the
attitudes of the respondents. Thus, one has to choose the timing of the retest in such a way
so as to strike a balance between the memory effect and the effect of changes in views.
Validity
The second attribute that a scale should have is its validity.
Validity is the success of the scale in measuring what it sets out to measure, so that
differences between the individual scores can be taken as representing true differences in
characteristics under study.
While the concept of validity is simple to understand, it becomes difficult to apply the test
of validity in practice. There are four approaches that can be commonly distinguished.
Content Validity
The content validity, implies that the contents of the scale correspond to the contents of the
attitude system and that they are comprehensive enough to cover the full range of the
attitude. The researcher should first define the problem clearly, identify the items to be
measured, and evolve a suitable scale for the purpose. Despite all this care, he may be
criticised on the ground that the scale developed by him lacks content validity. This happens
because, whether or not a given scale contains the content validity will depend on the
judgement of the researcher and, this is likely to vary from individual to individual. To
avoid this, it may be preferable to approach a group of knowledgeable persons, rather
than leaving it to one person alone.
Construct Validity
One of the difficulties arising in attitude measurement is that it is perhaps impossible to

measure attitude directly. It can be measured only indirectly on the basis of answers given
by the respondents. In a situation of this type, the test of construct validity is used.
The concept of construct validity is more complex than that of content validity. In order to
apply construct validity, the researcher postulates the nature and extent of the association
between the attitude and other specified variables. He then examines whether these
relationships exist. If not, there could be two possible explanations.
First, his scale might be invalid as it does not satisfactorily measure what it set out to
measure. Second, his theory might be deficient in some way and it may be difficult for him
to identify it. The point to emphasise is that the construct validity is based on theoretical
considerations. For example, the status of an individual in a society may be dependent
upon such variables as the level of education, occupation or ownership of a car and a
house. Thus, on the basis of theory, an elite class should have a high degree of association
amongst these factors. The existence of a high degree of correlation in this case is supporting
evidence and can be regarded as a test of validity.
Predictive Validity
Predictive validity signifies how best the researcher can guess the future performance,
from his knowledge of the attitude score.
For example, an opinion questionnaire which forms the basis for correctly forecasting the
demand for a product has predictive validity. The procedure for predictive validity first
measures the attitude and then predicts the future behaviour on the basis of this measurement.
Marketing Research
This is followed by the measurement of the future behaviour at an appropriate time. Finally,
the obtained scores are compared with the earlier predicted scores. If the two series of
scores are closely associated, the scale is said to have predictive validity.
Concurrent Validity
In the case of concurrent validity, an attitude scale on one variable can be used to estimate
scores on another variable. For example, one may decide the social status of respondents
on the basis of their attitude towards savings. Here, the attitude scale as also the criterion
measure is administered almost at the same time. It may be noted that a high degree of
concurrent validity may sometimes be spurious as the collection of one set of data may
influence the collection of another set.
General procedure in Attitude Scaling
Although there are a number of methods available for the measurement of attitude, the
most commonly used approach is the self-report, where a person is asked directly how he
feels about an object. The other alternative methods are observation of behaviour, indirect
techniques such as word-association tests, sentence-completion tests, story-telling,
performance of 'objective' tasks and physiological reactions. Since, the self-report method
of attitude measurement is most frequently used, the discussion here is based primarily on
it
One should assemble a set of items or statements related to the subject of enquiry. From
this pool of items, a final choice of items is to be made for inclusion in the scale. It is
necessary to exercise sufficient care so that complex, imprecise, vague or ambiguous
items can be avoided. Also, the statements should be written in simple language so that the
respondent easily understands them.
Having collected a set of items or statements, the next task is to choose the items for
inclusion in the final scale. By a process of elimination, unsuitable items should be discarded,
as a result of which the items left in the pool will form the final scale. While discarding the
unsuitable items, it should be ensured that the items retained are such as to comprehensively
cover the attitude dimension. For this purpose, an exploratory study is sometimes undertaken
in which some persons are asked to respond to all the items of the pool. Another approach,
as is used in the Thurstone scales, is to ask a group of judges to assess the suitability of
items for the final scale.
Finally, the scale, once formed, should be tested in regard to its reliability and validity. If
the scale satisfies these criteria, it can be used in the survey.
5.5 FORMS OF A RATING SCALE
The scales that are commonly used for attitude or opinion measurement are called rating
scales. Attitudes are mental states used by an individual the way they perceive and respond
to their environment. These are interval scales but are generally more complex and
multi-item in nature.
Rating scales are used in various forms: 1.
Itemised rating scale
These scales employ close ended questions with multiple choice answers on a
continuum that reflect a range of possible views regarding an object. These answers
are prompted and not spontaneous.
For example, show card.Please refer to this card and indicate how much you dislike
or like the taste of Bru coffee that you have just had.
Disliked very much Liked it somewhat

4 Liked it very much
1 5
Disliked some what
2
Neither disliked nor liked it
2. Diagrammatic/graphic rating scale
Is quite like the above scale except that the views may be scaled through a diagrammatic
representation. This may be used for illiterate persons, children etc.
Which of the faces best expresses your feelings on the coffee you have just had.
3. Rank order scale
This is a comparative rating scale in which a respondent is required to rank

comparatively the listed items.
For example, Questionnaire related to the job responsibilities
In your job you may perform these five duties. Consider how much time you will
spend doing each duty. Write 1 against a job that will be done most of the time and 5
against the job taking the least time.
Marketing Research
Coordination Forecasting
Problem solving Expert
guidance Absentee
replacement
This scale is successful if all possible sets of choices are indicated. Order bias and
items ranked outside respondent's preference set may render the response
meaningless.
4. Constant sum scale
It requires a respondent to allocate fixed numbers of rating points (usually 100) among
several attributes to reflect the relative importance of the attribute being studied.
For example, if you have Rs. 1007- to be allocated to various food items, how much
money will you allocate on each item?
Rice Milk
Wheat Biscuits
Ghee Tea
Total Rs. 100
5. Semantic differential scale
It is an insightful procedure for comparing the images of competing brand or services.

Respondents are asked to express their feelings on the attitude object on a five or
seven point rating scales bonded at each end by polar opposites. The central point of
the scale is a neutral point.
For example, Rate the TV programme "Kaun Banega Crorepati" for older people
and young people:
Scale
• VeryExciting
nn
I Unit 5
•
Scaling Methods
Somewhat Exciting
Neither Exciting nor Dull
• Somewhat Interesting
• Very Interesting
• VeryDull '
• Very Boring
Semantic differential scaling facilitates greater speed in ascertaining the response for
a range of brands across a battery of parameters. Problems of question phrasing and
structure are practically eliminated. It also provides a basis for isolating the intensity
of predetermined feelings. The only difficulty posed may be due to adjectives not
being precisely polar opposites or the 'halo effect', which tends the respondent to
mark all responses in one direction.
6. Likert scale
This scale was developed by Likert and hence the name. It is a five-point scale on
which a respondent is required to indicate a degree of agreement or disagreement
with a variety of statements related to the attitude object.
The first step in developing a likert scale is to generate a number of statements relevant
to the attitude.
These statements are given to a screening sample representative of the study population.
The next step is the item analysis, which chooses the subset which discriminates
among favourable and unfavourable attitudes. This is done by examining the correlation
between responses to each item and the total score obtained by summing up the
response to all other items. The most favourable and least favourable opinion groups
are retained and the mean is computed for each statement groups showing the biggest
difference between means.
For example, a study to evaluate Pune as a city to reside in the following questions may be
asked.
Ill
Marketing Research
Agree Neither Disagree

Agree Strongly Disagree
some what agree some
strongly
nor what
disagree
1. The weather is good
2. Is a safe place
3. Cost of living is
affordable
4. No language
problem
5. Good educational
institutes available
6. Good transportation
system
7. Getting domestic
help is easy
8. Sufficient medical
facilities available
The popularity of this scale is due to relative ease of construction and administration.""
Through this scale one can determine the favourability of disposition towards a statement
but to what extent may not be known. There is lack of reproducibility of the research.
5.6 SELECTING THE APPROPRIATE SCALE TO USE ___________

Before embarking on the scale usage, it is advisable to consider the following aspects.
1. Type / details of information required
2. Time and cost constraints

3. Literacy level of target respondents (Communication ease)
4. Stimulus characteristics
5. Analysis methodology to be used
».11
6. Past experience of appropriate scales

7. Rural v/s urban context (compatibility of scale structure to respondent's attitude)
8. Expected and desired response
Accuracy of attitude measurements
All measures used in marketing must be both accurate and useful. They must possess the
attributes of validity, reliability and sensitivity.
Validity -measuring what it is supposed to measure
Reliability- repeating the measurements within same parameters must yield a consistent
result
Sensitivity- the ability to discriminate among meaningful differences in attitudes
Though the foregoing definitions make it abundantly clear that defining 'attitudes' is a
difficult task, one does find that there is a common say running through all of them.
5.7 SUMMARY__________ . _________________________________^ _______
As we have seen in this unit, the different types of scaling are to be used by the appropriate
methods. We have also learnt what is the meaning of the word attitude, and further discussed
the types of scales, how to select these scales but we have to learn that scaling also has
limitations. In view to these limitations, scales should be used with great care and caution.
5.8 KEYWORDS ___________________________________________________

A
'*' Phenomenon: fact
Prejudice: preconceived opinion I

Preconceived: beforehand
- Convictions: being convinced

Marketing Research
6.1 INTRODUCTION
The difference between primary and secondary data and their advantages and disadvantages
will discussed in this unit. This unit will also help us to identify and select the primary and
secondary data as well as teach us the different scales to be used while designing the
questionnaire and also how sampling methods are to be used.
Data can be obtained in various forms. It can be categorised by the distinct difference in
meanings.
Facts: The measurement of anything that actually exists or has existed.

Knowledge: Knowledge is what people know. It is a determinant of how people act.
Opinions: Options are based on how people perceive something.

Intentions: Intentions are expected behaviour as one has in mind to do.
Motives: Motives are internalised factors that cause people to behave the way they do.
Data may be obtained through respondents. Respondents are individuals who provide
information through verbal, written or non-verbal clues (behaviour).
6.2 QUALITY OF DATA AND DATA TYPES____________________________

• The data objectives are derived from research objectives.
• The data accuracy is dependant on validity and reliability.
Validity - measuring what it should measure.
Reliability - repeating the same methods should produce similar results.

• The data must be obtained quickly enough at a reasonable cost.
6.3 DATA TYPES _________________________________________________

The different types of data are illustrated below:
Demographic Sociological Economic Psychographic Behavioural

Pertaining to How people Pertaining to Individual mental How people
population are organised financial makeup, lifestyle actually
in a society strength etc., behave
118
Unit 6 Data Collection Primary / Secondary
6.4 SOURCES OF DATA

There are basically two sources for data namely
Primary: comprises consumers, industrial users, middleman and other knowledgeable

people.
Secondary: comprises internal and external sources. Details are shown in this unit further.
6.5 COMMUNICATION MEDIA

As simple division of the communication media is as
Manual
follows: Observation Structured v/s Unstructured v/s
Mechanical
Questioning Disguised v/s Known
Experimentation
Communication media decision should be taken keeping the following factors in mind.
1. Freedom from interviewer bias
2. Sampling size
3. Speed of response
4. Versatility of the mode
5. Rapport with respondents
6. Cost of operations
7. Data collection control
8. Indepth information
9. Hard to recall data tapping
10. Follow up required

Marketing Research
6.6 FORMULATING A DATA ACQUISITION PLAN
Set
Data
Objective
Select the
Appropriate
Data Type
Define the Source

of Data
Select the Communication

Approach Mail, Telephone, Personal,
Observation
Data
Collection
6.7 BASIC METHODS OF GATHERING PRIMARY DATA _____________
Primary data is the original data that a researcher gathers for a particular project. There
are three methods of primary data collection.
• Observation
• Experimentation
• Surveys
Direct ->•
1. Indirect
Observation Mechanical
a) Direct Observation
It is a process of recognising and noting people, objects and occurrences rather

than asking for information in a natural situation. For example, number of vehicles
at a road junction, grocery shopping of individuals, etc.
b) Indirect/Contrived Observation ''
At times, researchers feel that a lot of time is going to get wasted to wait for a
desired event to happen. So artificial stimulus situations are created and the
respondent's response is studied. For example, in the pharmaceutical industry,
to see how a medical representative is going to deal with a doctor, artificial
clinics and doctors are placed. The sales call of a representative is recorded
without his knowledge to study his call effectivity and to suggest methods to
improve his call.
c) Mechanical Observation
In the above two methods, the human interface is there. Newer technology
entails the use of mechanical devices for observation. For example,
1. Audiometer: to record radio and television timings when they are tuned
on.
2. Web camera: to check in shop mobility, shop lifting etc.
3. Eye camera: records movements of the eyes.
4. Psycho galvanometer: measures minute emotional reactions through change

in perspiration just like a lie detector.
here
2. Experimentation
It is difficult to apply this method in actual practice, as it is a tightly controlled method.

A select stimulus is applied under a controlled environment to study cause and effect
relationships.
3. Surveys
Questionnaire and various specialised techniques (for example, shop audits, consumer
panels etc.) are used to collect primary data.
4. Mail
Postal surveys are carried out by mailing self-administered questionnaires to the desired
ither respondent instead of an interviewer asking questions. Due diligence must be exercised
icles while designing such questionnaires for a proper response.
Marketing Research
6.8 ADVANTAGES AND DISADVANTAGES OF PRIMARY DATA

The advantages and disadvantages of Primary Data may be said to be as follows:
Advantages Disadvantages
1) Sample size may be dispersed and Incorrect postal addresses and

distant areas and inaccessible an inadequate mailing list.
persons may be tapped.
2) Respondents may respond at a It may take weeks before the

time convenient to them without any responses start coming back.
pressure.
3) Confidential information may be Since a respondent has read

divulged if identity is concealed. the question sequence, there
may be an answering bias.
4) Using reply - paid envelopes is Apparently the original low cost

cheaper, taking into consideration becomes high, due to a poor response
interviewer's training and rate. Some respondents may not
transportation costs. respond at all.
5) Interviewer bias is avoided It may not have been filled by the

respondent himself. Therefore it
will not represent the true
sampling universe.
Telephonic Interviews
Respondents are reached for surveys/interviews through telephones. The usual methodology
of sampling and designing of a questionnaire is followed except that responses are filled by
the interviewer on telephone.
1) Coverage of a widely scattered Only respondents having a

sample is rapid. telephone may be contacted.
2) Low cost in comparison to door Visual aids cannot be used for

to door personal interviewing. interviewing.
3) Frankness of views as against Lengthy interviews not possible.

written communication. Respondents get suspicious as
the interviewer cannot be seen.
Technical errors marr

4) Greater speed of contact since functioning of telephones.
distance is not a barrier.
Respondent communication
5) Better supervision and control problems cannot be visualised.
over interviewers. Retaining attention for long is difficult.
Personal Interviews
This is the traditional method of data collection and can take place at any location. The
respondent is face to face with the interviewer to answer the questions.
Structured v/s Unstructured

In structured interviews, predesigned and preset questions are asked through a questionnaire
to the respondent whereas in unstructured interviews, the interviewer covers varied aspects
of the subject at length. So sometimes they are referred to as indepth interviews.
Traffic Interviews
Interviews are held on streets and at public places.
Mall Intercept
Targeted respondents are intercepted at shopping areas.
Group Interviews
A questionnaire is administered to a homogeneous group of people who are shown/explained
certain facts and are then required to respond to a questionnaire. For example,
demonstration on a microwave and a questionnaire on a particular brand to housewives.
This may serve the purpose of research and sales promotion at the same time.
1 . Flexibility - audio, visual and mechanical It is a time consuming and

aids may be used to conduct interviews. expensive exercise.
123
Marketing Research
2. The interviewer may persuade the Staffing capable interviewers and

respondent to answer the questions. controlling them is a problem area.
3. Comprehension difficulties may be Interviewer's bias may influence

sorted out. responses.
3. Detailed information collection is possible Respondents are reluctant to talk to

Certain facts can be observed without strangers.
questioning.
4. Selection of sample size can be more Scattered and distant sampling area
precise. is difficult to cover.
6.9 TYPES OF SECONDARY DATA COLLECTION AND ITS

ADVANTAGES & DISADVANTAGES
Internal xternal
Published Unpublished
Studies
Balance sheet Sales Reports Unpublished Research
Director's Report Purchase/Sales Government Scholars
Auditor's Report Profit invoices Publications Municipalities
& Loss Account Challans (census/ministry
Chairman's Statement Budgets reports))
Annual Report News Research Reports Association Reports
Letters/Inhouse Departmental International Agencies
magazine communications (UNO, IMF)
Professional Bodies
(FICCI)
Subscription Services
Libraries (Chamber of
Commerce)
Indian Archives
(Pre Independence Era
Historical facts)
Unions (Sugar. Trade)
Newspapers, Journals
Periodicals
Stock Exchange Reports
Web pages/Online Data
Services
Handbooks/Encyclopedia
Indexes
(e.g. Stock Exchange)
Commercial Research
Services (Audit Bureau of
Circulations)- ----- •—
laeological Survey of India
ps)
Export Promotion Councils
And Boards
Secondary Data
1. Company Audits, Records and Annual Report: Every limited company publishes
an annual report regarding its progress for the stakeholder, which is an indicator of
the general performance of the company. Also, there may be an audit on various
issues like inventory control, working capital, fund utilisation, etc. which are of a
great help to a researcher. Monthly records on sales, raw material usage, the profit
and loss statement and the budget are also unpublished internal sources of information.
2. Government Publication: The government publishes books, official gazette, census,

journals, industrial guidelines, etc. as references. Some of them are given below.
Publications
Sr. No. Name Publishers
1. Architecture All India Architects Directory
2. Arecanut Indian Arecanut, Spices & Cocoa. Journal of Plantation Crops
4. Cashew Cashew Bulletin, Indian Cashew Journal

5. Ceramics Indian Ceramics, Central Glass & Ceramic
Research Bulletin
6. Chemicals Indian Chemical Directory
Catalogue of Indian Chemical Plant Mnufacturers
Chemicals & Petro Chemicals Journal Chemical
Industry News, Chemical Industry Development
Marketing Research
7. Export/Import Export-Import News, Indian Engineering

Exporter, Daily Bulletin of Exports
8. Footwear & Journal of the Leather Technologists

Leather Association
9. Silk Indian Silk, Indian Journal of Sericulture Silk in India
10. Spices Spices News Letter, Indian Spices

11. Wool Wool and Woollens of India
Research Institutes
1. National Research Development Corporation of India, New Delhi
2. Council of Scientific and Industrial Research, New Delhi
3. Building Research Institute Central, Roorkee
4. National Chemical Laboratory, Pune
Central Government Undertakings
1. Agricultural Co-op. Marketing Federation of India Ltd., (NAFED), New Delhi.
2. Agricultural Finance Corpn. Ltd., Bombay
3. Agricultural Refinance Corpn. Bombay
4. Banana & Fruit Development Corpn. Ltd., Madras
5. Bharat Leather Corpn. Ltd. Agra
Export Promotion Councils/Commodity Boards
1. Basic Chemicals, Pharmaceuticals & Cosmetics Export Promotion Council, Mumbai
2. Cashew Export Promotion Council, Chittor Road, Cochin
3. Chemicals & Allied Products Export Promotion Council, Kolkata
4. Cotton Textile Export Promotion Council Engineering Centre, Mumbai

5. Engineering Export Promotion Council, Kolkata
Advantages and Disadvantages of Secondary Data '
1. Easy Accessibility
2. The cost of conducting a full scale research is not incurred, so it is more economical.
3. It is time saving, as information may be availed without doing actual field surveys
within a few days.
4. They provide for information, which may not be availed by a typical organisation e.g.
The Bureau of Census and the sales figures of retailers.
5. The data is not gathered for an organisation with specific purpose; hence researcher's
bias is minimised.
6. Helps in problem definition, the formulation of hypothesis and planning the data
collection.
Drawbacks of Secondary Data
1. Unit of Measurement
Different data agencies use different units of measurement. For example, expenditure
census may use expenditure per household whereas an income tax department takes
it on income at source on individual basis as the basic unit for measurement.
2. Recency
.3- Certain data is collected after a lapse of five or ten years which is a long period for
marketing purposes. There may be a lot of factoral changes during the said period
and this may render the information quite obsolete. For example, the Census of
population.
3. Definition of classes
Classes are made by individuals to suit their research requirements. For example,
youth may be defined as all people below 30 years in one research while another
reference may classify it as all people below 40 years. For people who are using
cross references for their purpose of study may find it very difficult to tabulate and
use such varied sources of information.
127
Marketing Research
4. Inaccuracy
Arithmetical errors, clerical errors, printing errors, omission of footnotes, cautionary

notes and explanatory notes may render the data inaccurate.
5. Incompleteness
Very seldom the data available embraces the entire population. A DGTD report on
confectionery manufacturers may give details about the organised units reporting to it
while the small scale units in the unorganised sectors may be inadvertently ignored.
6. Assumption"
Each research takes certain variables as a constant for the purpose of study. Aspects
such as price rise, devaluation, tax structure changes etc., have their own impact on
the economy and may reflect a partial picture.
7. Source of data
While evaluating the secondary data, a researcher must examine the source of data,
the data collecting agency and the purpose of publication. For example, income tax
department will give more accurate data on income level since it is authorised to
collect information rather than a private organisation giving this information.
8. Misleading
The statistical data is sometimes misleading. For example, a report stating the annual
production of electrical motors at 110 million H.P. The share AC, DC, various range
of HP is missing in it.
6.10 RESPONDENTS AND THEIR DESCRIPTORS

The difference between research context and the target respondent is given below :
Research context Target respondent

1. To ascertain the experience of Users of the product
2. Pro-V hair conditioner Users of other conditioners
Reasons for non acceptance of Non-users of hair conditioners
Pro-V hair conditioner
3. To understand the knowledge Information could be got from

and attitude towards hair Any sex-male/female
conditioners Age-Specification Household
-type by income Education
level Urban/Rural
background Town
specification
In the first two contexts, the profile of the target respondent is fairly simple but the third
context renders the profiling rather complex. Not all individuals can meet the selection
criteria which describes the individualneeded for the study. Such describing criteria are
called respondent descriptors.
Types of Respondent Descriptors
Geographic Town
+ r i
zones class
;
Demographic
f +
Age North Urban User Primary MHI Attitude Changed
Sex East Rural Non user Secondary Education Interests Lifestyles
Monthly West Opinions
household South
Income
Marital status
household
structure
Education
Occupation
Working status
Religion
Basic food
habits
Mothertongue •
Demographic: The use of demographic descriptors depends on the product category

and the research context. Generally, a combination of variables is used to define the target
audience. For example, age, sex and income criteria may be used for one study.
Standardisation of descriptors helps in a comparative study of various researches, cross
tabulations and building weightage.
Questions may be asked to get response from descriptors for example,
Age: What is your age
Marketing Research
Sex: Male/Female - to be ticked by observation or filled by the respondent

Occupation : Show card (A handout card containing various options is shown to a
respondent).
Please refer to the card and select the phrase that best describes your occupation.
Some descriptors
Sex
Male
Female
Age
Scale intervals
0-5 years 6-10 31 -35 years
can be made as
years 11 -15 36 - 40 years
per need of the
years 16 - 20 41-45 years
study.
years 21-25 46 - 50 years
0-5 years 31-35 years
years 26 - 30 51-55 years
years 55 + years
Not disclosed.
Marital status
Single / Never married
Married Widowed
Separated/Divorced
Not disclosed
Working status
Networking -
unemployed
housewife
student
retired
Working part time (< 4 hrs/day)

Working full time (> 4 hrs/day)
Not disclosed ^
Occupation
Unskilled worker
Skilled worker
Petty trader
Shop owner
Business/industrialist (can be classified further by the number of employees)
Self-employed professionals
Clerical/Salesman
Supervisors
Officer/executive (Junior)
Officer/executive (middle/senior)
Not disclosed
Household structure
Nuclear family
Nuclear with elders
Jointfamily
Single person/siblings/others
Not disclosed
Education
Illiterate
Literate without formal schooling
School - upto 5 years (primary)
- 6 to 9 years
-SSC-HSC
Marketing Research
Diploma
Graduate/Postgraduate (general)
Graduate/Postgraduate (Professional)
Not disclosed
Mother Tongue (Listed Alphabetically)
Assamese Sindhi
Bengali Tamil
English Telugu
Gujarati Urdu
Hindi Others ______
Kannada Not disclosed
Malayalam
Marathi
Oriya
Punjabi
Family Life Styles
Unmarried
Married with no ch i Idren
Married with young children (below < 12 years)
Married with teenagers
Married with college going children
Married with married children
Married with children not staying with them
Religion
Hindu
Unit 6 Data Collection Primary /
Secondary
I
Sikh
Jain
Muslim
Christian
Buddhist
Others
Not disclosed
Household Income
Rs.350-Rs.500 -Lower
Rs. 501 - Rs. 1000 - Lower middle
Rs. 1001 - Rs. 2000 -middle
Rs. 2001 - Rs. 4000 -Upper
Rs. 4001 - Rs. 10,000 - Upper middle
Rs. 10,0001+ -Upper
Not disclosed
Food Habits
Non-vegetarians
Vegetarian
Not disclosed
Eggitarians (vegetarians eating eggs)
Geographic: India is a diversified country with a lot of regional differences. For a research
zone (east, west, north south, state level, district level), a descriptor could be made. The
state reorganisation commission (1954) has set up districts and state boundaries based on
the homogeneity of socio-cultural parameters. These parameters include:
• Language homogeneity
• Geographical continuity
• Financial, economic and administrative homogeneity
• Regional and cultural similarities
133
Marketing Research
Caste and Class Consideration
Town classification Village classification (Rural)

Classification Population Classification Population
Metro < 10 Lakh 5-10 Class I greater than 5,000

Class I Lakh 1 - 5 Class II 1000 - 5000
Class II Lakh 50,000 - 1
Class III Lakh less than Class III less than 1000
Class IV 50,000
Owner of products / service
Users Non-users
Typical example
An individual can be a user/non-user of a particular product or service if he/she uses

certain other products.
Primary Secondary
Airlines Frequent flier Programmes
Microwave Microwave dishes
Socio Economic Class
Earlier monthly household income (MHI) indicator was used to study purchase behaviour
in India. Now the market research society of India has developed a hybrid of two descriptors
- the education and occupation of the chief wage earner in a household. Each member of
the household will have the same SEC. This is called as the socio-economic classification
(SEC). According to NRSIV, the median MHI value are given below.
SEC Median MHI Value (in Rs.)

A1/A2 3180
B1/B2 1900
CD 1440
910
El 780
E2 620
Rural SEC Grid: It is based on education and Pucca/Semi pucca/Kuchha house.
Psychographic: It includes activities, interests, opinions of people which keep changing

over a period of time. Lifestyle factors are also considered.
For example, I want a car to make my son happy I take pride in
cleaning my car Educating your children is a must I like listening to Jazz
Politicians are big scoundrels Various scaling and projective techniques
help to measure these factors.

Psychographic: A specific change in lifestyle is indicated through psychographic. It depicts
how people allocate their time and money to certain products and factors. For example, a
person starting his first job, getting married, changing residence, etc. Depending on the
research context, psychographic descriptors may be used. For example, a baby care
product. These are a few listed descriptors. Aresearcher may alter and add descriptor to
suit the research need.
6.11 SUMMARY ________________________________________________

Market research involves collection and analysis of data. Data may be collected through
primary or secondary sources. Without data, research is impossible.
Primary data involves interaction with individuals who provide the information. These
individuals are called respondents.
Before approaching any individual, a researcher needs to decide on the profile of the
resi ipondents needed for study. This is the target audience. The profile of the target audience
will depend on the nature of the research.
135
Marketing Research Unit
6.12 KEYWORDS Qut

Tabulations: arrange in tabular form
Qi.
Nuclear family: couple and their child or children
Q2.
6.13 CASE STUDY Q3.
A leading group of Hotels •*
Aleading group of hotels owns a chain of 20 hotels located in different parts of the country. 6.14
In recent years, it has been expanding the chain by setting up new hotels.
Q.i
When there were few hotels, the top management of this group of hotels used to personally
visit them with a view to ensuring that they provided high quality food and services to their
patrons. But, now they find that with so many hotels it is extremely difficult to personally Q.2
visit each and every hotel. At the same time, they need some mechanism to ensure that
hotels of the chain continue to provide high quality service.
Q.3
The Management has discussed this problem with some senior officials of the company.
As a result of this discussion, the management felt that out of the options given below
which option would be suitable for their group and how it will help them to monitor their
parameters.
1. A suitable questionnaire may be designed and the same may be given to the patrons
during their stay in the hotels. They may be requested to return the filled-in
questionnaire at the reception counter while checking out of the hotel.
2. A suitable questionnaire may be designed and the same may be posted to their
residence soon after they have reached there with the request that these be returned,
duly filled in by post.
3. A trained interviewer may be appointed. He could visit different hotels without any
prior intimation of his visits. In each hotel he visits, he may personally interview selected
guests and seek their opinion on the quality of food and services in that hotel and their
suggestions for improvement if any.
4. The services of a marketing research firm may be hired. It may be asked to conduct
a suitable study based on, say, telephone interviews of a random sample of guests
from each hotel and to submit a report to the Management.
Questions
Q1. In your view which option is to be adopted?
Q2. Discuss the relative merits and demerits of each option?
Q3. If you are the Chairman of the group, which is the best method you will adopt and
why?
6.14 SELF-ASSESSMENT QUESTIONS _______ _ _ ________ '

Q. 1 An NGO wants to establish a charitable hospital in the town. They cannot ascertain
which specialties to start. Formulate a data acquisition plan to resolve their problem.
Q.2 A mail survey would be less expensive than a personally administered survey.
Comment.
Q.3 The utility of secondary data cannot be ignored. Comment.
137
Marketing Research
7.1 INTRODUCTION
Once the data is collected, the process of analysis is triggered off. The raw data which is
received from the field cannot be used as such for interpretation unless the validation of
data has taken place. Seemingly small errors can lead to misleading results.
7.2 TRADITIONAL APPROACH OF DATA PROCESSING_____________

The traditional approach of data processing is simplified with the help of the following flow
chart:
Validate data
I Edit acceptable
questionnaires
!
Code
I
Transcribe
1
Clean data
I Store data for
analysis
!
Select an analysis strategy
1. Validation: Determination of data validity is a fundamental step in research. Validation

ensures that data has accurately measured what was intended to be measured. Two
steps of validation are generally done.
• Sampling-for cause and location error
• Questionnaire-errors in the conduction of interviews
2. Editing: It is the process of review of questionnaires with the objective of increasing

accuracy and precision. The following errors are generally detected.
• Fake interviews
• Inadequate answers
• Incomplete answers
• Ambiguous responses
• Inconsistent response
Unit 7 Data Preparation and Processing
• Illegible responses ' '

• Irrelevant answers
These errors are rectified by contacting the respondent if possible, by trying to infer
meaning by subsequent answers or discarding them altogether.
3. Coding: Coding is the procedure of classifying the answers to a question into meaningful
categories. The symbols used to indicate these categories are called codes. Coding
is necessary to carry out the subsequent operations of tabulating and analysing data.
If coding is not done, it will not be possible to reduce a large number of heterogeneous
responses into meaningful categories with the result that the analysis of data would be
weak and ineffective, and without proper focus.
Coding involves two steps

a) The first step is to specify the different categories or classes into which the responses
are to be classified.
b) The second step is to allocate individual answers to different categories.
Code construction, as pointed out is something of an art and the final categories reflect the
tastes and interests of the individual researcher. All the same, it is desirable to follow some
guidelines to ensure the utility and rationality of the code. One of the most important points
in this respect is that the categories must be all inclusive and mutually exclusive.
The 'all-inclusive' aspect can be taken care of by adding one or more such categories as
'other', 'no information' and 'none'. The other aspect is that categories must be 'mutually
exclusive', i.e. they must not be overlapping and ambiguous. It should be possible to
classify each response in one and only one category. However, this requirement is often
violated when more than one dimension is embodied in a single case. To give an example,
a person may, by occupation, be an industrial worker as well as unemployed. Here, two
concepts or dimensions have been used. The first is the occupational category and the
second is the current employment status. In such a case, there is apprehension that different
categories or classes will not be mutually exclusive. It would, therefore, be advisable to
use two category-sets, one for the occupation and the other for the current employment
status.
There is no definite rule for the number of categories or classes that can be used. This will
depend on the research problem as also the extent of analysis the researcher proposes to
carry out.
141
Marketing Research
In large surveys, where mostly structured questionnaires are used, the response categories
are pre-determined and are contained in the questionnaires themselves. The categories
are in the form of multiple-choice answers to the question.
For example, the respondent may be asked: To which age group do you belong? 15-30
years, 30-45 years, 45-60 years and 60+ years. Here, four distinct categories are indicated
and the respondent is supposed to indicate his by checking the category in which his age
falls. It is obvious that in such a case the respondent himself chooses the category which is
applicable to him.
The problem of coding is not so simple, especially in respect of an open-ended question.

The response to such a question is in a descriptive form, in the words of the respondent
himself. For example, the respondent may be asked:
What is your opinion regarding the prohibition policy of the government? The respondent
may give a lengthy answer indicating what he feels about this policy. In case of such
responses, coding needs extra care in framing the possible categories in which various
responses can be classified. Sometimes the interviewer himself decides the category in
which a particular response to an open-ended question is to be included. He may first take
down the entire response and then decide the category in which it should be included.
At times, the questionnaires are wholly or partially pre-coded. These questionnaires contain
a numeric code for each of the response categories.
Some examples of pre-coded questions are given below:
Questions Answers Codes

How often these days you watch your More than once a week 12
favourite TV program? Once a week Once a 34
fortnight Once a month 56
3 to 4 times in a year 71
Less often 23
Never
Hand wound
Which type of wrist watch you have? Automatic
Quartz
142,
Which of these equipment do you have Microwave 1

in your family? Television Washing 2345
Machine Dish 67
washer CD Player 89
Digital Camera
Lap top
Personal computer
Any other (specify)
Sometimes data are transferred from the questionnaire to a coding sheet and then punched
into cards.
A practice which is frequently followed is to edit and code the data simultaneously. These
two operations are regarded as one operation which is looked after by one person. It has
been rightly pointed out, although this may perhaps be the quickest and most efficient
method, it may lead to the neglect of editing as the editor who is expected to code becomes
just a coder. In view of this, it may be advisable to get these jobs done by two people.
However, in such a case, coding by itself tends to be monotonous and boring but this
perhaps cannot be helped.
In simple terms, coding means assigning a code usually a number to each possible response
to the question. This code could be in letters as well. Coding enables transferring of data
in a computer readable form. Tick the appropriate number
e.g. Male 12
Female
The respondent code and record number should appear on each record of data. Additional
codes consist of project code, interviewer code, date code, time code and validation
code.
m
7.3 PRINCIPLES OF CODING _______________________________________
The principles of coding can be said to be:
• Convenient number of categories should be there so as to reveal differences in data.
• Category codes should be mutually exclusive and collectively exhaustive.
143
Marketing Research
• Category codes should be assigned to critical issues even if the category is no response.
• Class interval if possible should be of same width.
• Coding unstructured or open ended questions gets complex and hence should be
avoided.
• Coding should reflect inter-class homogenity and intra-class heterogenity.
Codebook: Codebook is a guidebook for researchers to help them identify and locate
variables. Sometimes, instead of writing codes on the questionnaire itself, the codes are
written on a special 80-column paper called the coding sheet.
4. Transcribe: Transcribing data involves transferring the coded data from the
questionnaires or coding sheet onto disks, magnetic tapes or directly into computers
by key punching. Data can be optically scanned and involves direct machine reading
of codes and simultaneous transcription e.g. Product bar code (universal product
code).
5. Data cleaning: It is done through extensive checks for consistency and the treatment
of missing responses. Consistency check identifies data that are out of range, logically
inconsistent or have extreme values. Missing responses denote unknown variables,
ambiguous answers or those not recorded properly. This problem is resolved by
substituting a neutral value or an imputed/ calculated response.
6. Data storage: With the advent of the microchip, computer aided technology has
created a revolution. Huge amount of data can be stored and retrieved from a
microcomputer in no time. Various statistical software help in easing computing
problem.
7. Selecting an analysis strategy: Interpretive goals spell the requirement for data
processing and statistical analysis. One needs to select a method out of the four data
processing approaches:
• Manual or personal method
• Data-processing experts
• Use of data analysis programs ?
• Develop computer software as per needs

As regards the choice among the four alternatives, the trend is towards electronic rather
than manual processing. Existing data analysis programs may be used if they cover the
marketing program. Thus, starts the actual task of analysis.
Once the data has been collected, the researcher has to process, analyse and interpret the
same, hi unit 6, it was emphasised that the researcher should exercise good care to ensure
that reliable data is collected. All this effort, however, will be in vain if the collected data is
not properly processed and analysed. Sufficient attention is often not given to these aspects,
with the result that the quality of the report suffers. It is desirable to have a well thought-out
framework for the processing and analysis of data prior to their collection.
Dummy tables should be prepared in order to illustrate the nature and extent of tabulation
as also the comparisons of data that will be undertaken. At the same time, it may be noted
that certain changes in such a framework may become necessary at a later stage. The
researcher should not hesitate to introduce such changes as may be necessary to improve
the quality of tabulation and the analysis of data.
7.4 EDITING __________________________________________________

The first task in data processing is the editing. It is the process by which data is prepared
for subsequent coding. As it is a very subjective process, it is necessary that people who
are well-qualified and trained in the j ob of editing, should alone be entrusted with this
responsibility.
Editing is the process of examining errors and omissions in the collected data and making
necessary corrections in the same. This is desirable when there is some inconsistency in
the response or responses as entered in the questionnaire or when it contains only a partial
or a vague answer. Given some few examples will indicate how editing can be helpful.
The respondent has given answers which are inconsistent with each other. In such a case,
the editor has to change one of the answers so as to make it consistent with the others. He
has to use his judgement to decide which answer is correct so that the other one can be
suitably changed.
The respondent has marked two answers instead of one for a particular question.In such
a case, the editor has to carefully examine which of the two answers would be more
accurate. Sometimes, when a decision cannot be made categorically, he may prefer to
code 'no information' for that question.
The respondent has answered a question by checking one of the many possible categories
145
Marketing Research
contained in the questionnaire. In addition, the respondent has written some remarks in the
margin. These remarks do not go well with the particular category marked by the
respondent. The editor has to look into this and may have to change the category to better
represent the remarks made by the respondent.
Sometimes the questionnaires contain imaginary and fictitious data. This may be due to
cheating by the interviewers who may fill in the entries in the questionnaire without actually
interviewing the respondent. This may also happen in the case of a mail questionnaire,
where the respondent has given an arbitrary answer without exercising any care. If the
responses indicate obvious inaccuracy, they may be either dropped or suitably modified if
they are to be retained. The editor has to exercise his judgement in this regard.
In all cases where editorial corrections are to be made, it is necessary that these should be
kept distinct from the changes made either by the respondent or by the interviewer. This
can be ensured by the editor by using a different coloured pencil for editing the raw data.
Editing can be undertaken at the time when the field survey is in progress and when it has
been completed. In the former case, it is known as field editing. When the interviewer fills
in the information at the time of the interview, he often uses several abbreviations due to
the paucity of time. These need to be spelt out fully later. It is advisable for the interviewer
to carefully look into the questionnaire at the earliest possible opportunity after the interview
so that he can enter the proper responses or complete the partial answers.
Another type of editing is central editing, which is undertaken after the questionnaires have
been received at the headquarters. As far as possible, a single editor should carry out this
task so that consistency in editing can be ensured. However, in the case of large studies,
this may not be physically possible. When two or more editors are entrusted with the task
of editing, it is necessary that they are given uniform guidelines so that the maximum possible
consistency in their approaches can be attained. An alternative way would be to split the
entire task into two or more parts so that each part of the work can be looked after by one
single editor. In such a case, chances of inconsistencies pertaining to the responses of a
particular part can be almost fully avoided.
An editor should be well-versed with the editing of questionnaires. It may be emphasised

that editing a manuscript is different from the editing of a questionnaire or numeric data.
People who are good at editing descriptive material may not be able to edit numeric data
satisfactorily. Persons who are quite efficient in detecting flaws or errors in the data in just
one glance should be entrusted with this job. Enumerators with long experience and having
a special aptitude for the editing of data should be given preference over others.
When the services of more than one editor are required, it is advisable to give each one
explicit editing instructions in order to ensure consistency in the editing of data.
Before undertaking the coding, tabulating and analysis of responses contained in

questionnaires, they should be checked for completeness, accuracy and uniformity.
The first point to check is that questionnaires are complete and do not have any omissions
or partial responses. Sometimes^the interviewer might have forgotten to record the answer.
In such cases, it may be difficult to fill in the gap as the interviewer may be unable to
recollect the answer given by the respondent. When several questions remain unanswered
in a questionnaire, the whole questionnaire may have to be excluded. However, before
doing so, all other alternatives should be exhausted and the questionnaire should be disposed
of only if it is unavoidable.
The second point to check is that questionnaires contain accurate answers. The editor
should look for inconsistent answers, which are sometimes so obvious that a little careful
perusal may detect them. Such inconsistencies should be removed. Sometimes inaccuracies
may be a result of the carelessness of the interviewer who may mark a wrong code or put
the mark in such a way that it is not clear which of the two codes is intended. At times the
respondent may give wrong or misleading answers deliberately. In all such cases the editor
has to go over the answers carefully and try to remove the inconsistency in the best possible
manner.
Finally, one should check whether the interviewers have interpreted questions and
instructions uniformly. While going through the questionnaires filled in by different
interviewers (or respondents in case of a mail questionnaire), the editor would be able to
make out such inconsistencies.
In dealing with these three points—completeness, accuracy and uniformity, the editor
should see that far too much time is not spent on trivial or relatively minor errors.
7.5 TABULATION _____________________________________________

Tabulation comprises sorting of the data into different categories and counting the number
of cases that belong to each category. The simplest way to tabulate is to count the number
of responses to one question. This is also called univariate tabulation.
The analysis based on just one variable is obviously meager. Where two or more variables
are involved in tabulation, it is called bivariate or multivariate tabulation. In marketing
research projects, both types of tabulation are generally used.
147
Marketing Research
The tabulation may be done by hand or by machine or some part by hand and the other by
machine. The number of tabulations will depend on the number of variables, while the
number of responses to a question will depend on the sample size. If the number of variables
in a survey is not large and the sample size is small, then tabulation by hand is preferable.
On the other hand, when the number of variables involved is large as a result of which
there may be several relationships and the sample size is large, it will, be more appropriate
to opt for machine tabulation.
Hand tabulation
Normally, hand tabulation is carried out with the help of a tally sheet. Consider, for example,
the following question:
How many movies did you see last week? For a sample of say 60 respondents, the
tabulation may be as shown below:
Table 7.1
No. of Frequency
movies seen
0 imiiiii 9
1 iiiiiiiiiiiiiu 15
2 miiiiiiiiu 13
3 iiiiiiiin 10
4 5 or more iiiiini 8
iiiii 5
60
The hand tally can also be used for preparing cross-classification tables. Consider, for
example, the relationship between the economic status and the number of movies seen.
It may be noted that the total column at the extreme right in the table below (table 7.2)
contains the same figures as shown earlier in the case of straight tabulation, as in the case
of Table 7.1.
Table 7.2
No. of Movies seen Rich Middle Poor Total
0 - iiii iiiii 9
~
1 limn iiiinii 15
2 11iii 1111 limn 13
»3 iii iiii 10
4 5 or i iI iiiiii 8
5
more iiii
60
Machine tabulation
In the case of large scale surveys, where a good deal of data from a large number of
respondents has been collected, hand tabulation will be bothersome and take much longer
to complete. In such a case, the method of machine tabulation is used. For this purpose,
mechanical sorting and tabulating equipments available with such concerns as the Hindustan
Computers Ltd and Remington Rand are used. These equipments include key punches,
sorting machines and tabulating machines. The main advantages of mechanical tabulation
are:
Extensive and large-scale survey can be handled conveniently.
It takes much less time than hand tabulation.

Greater accuracy in the sorting and counting is achieved with the help of machine
tabulation.
Monototly and drudgery of tabulation can be avoided by passing on this work to the
machines.
In the case of large-scale tabulation, mechanical tabulation eventually turns out to be
cheaper than hand tabulation.
Cross-tabulation, depicting the inter-relationship between two or more variables can

be undertaken easily.
Marketing Research
One-way tabulation
A one-way table, from the first hand tally given earlier Table 7.1 while Table 7.2 shows
both, the absolute frequencies and the percentage of respondents. Although it may not be
necessary to give percentages, it is a good practice to include them, as they facilitate
comparisons.
The researcher should ensure that percentage figures add up to 100. He has also to decide
up to what decimal place percentages should be given. Generally, figures may be given
upto one decimal place. Rarely, if ever, do they need to be given to two decimal places.
While such figures would be more accurate, they would also be more confusing to the
reader. The guiding principle in reporting percentages is that unless decimals serve a useful
purpose, they should be avoided.
Sometimes only percentages are shown in the table and the corresponding frequencies are
omitted. In such cases, it is necessary to indicate the total number of cases on which the
percentages are based.
Movies seen by sample respondent Table 7.1

No. of movies seen per week Number of respondent % of respondent
0 9 15
1 ' 15 25
2 13 22
3 10 17
4 • 8 13
5 5 8
60 100
Cross tabulation
It may often be necessary to tabulate responses to two or more questions simultaneously.

Such tabulations are known as bivariate or multivariate tabulations, depending on whether
two or more than two variables are involved.
In constructing cross classification tables, one has to first determine which data should be
given primary emphasis and which should be given secondary emphasis. Data with primary
emphasis are normally given in columns while those with secondary emphasis are shown in
rows. This order is repeated for higher order tables, i.e., those having three or more
dimensions. This convention is almost invariably followed because it is easier to see data
when figures follow one another in a column rather than in a row.
Table 7.3 : Is an example of cross-classification

Income Level Shopping Centre Shopping Centre Total no. of
house holds
A B
UptoRslOOO 100 200 300
Rs 1000 and above 120 80 200
Total 220 280 500
Table 7.3 gives the break-up of respondent households, both by income level and by their
preference for shopping centres. A table of this type is also known as contingency table.
This is the simplest contingency table with two rows and two columns. The data contained
in Table 7.3 poses an important question — does the preference for shopping centres
depend on the income level of households? To answer this question, it is necessary to
analyse the two variables simultaneously.
Data shown in Table 7.3 can be transformed into percentages and then these percentages
alone can be shown or they can be shown side by side with the original data. The question
in a two-way tabulation is that which base should be used for 100 per cent, as the data
maybe percentaged in either dimension. Percentages should be based on totals of rows
or columns, whichever is relatively more important. One simple rule in this regard is the
cause and effect rule. ,
This rule states that the percentages should be computed in the direction of the causal
factor. Table 7.4 and 7.5 give these data.
Table 7.4: Income levels and preference for shopping centres %

Income Level Shopping Centres
A B
UptoRslOOO 33 67 100
Rs 1000+ 60 40 100
Marketing Research
Table 7.5 : Income levels and preference for shopping centres %

Income level Shopping centres A Shopping centres B
Up to Rs 1000 45 71
Rs 1000 and above 55 29
Total 100 100
In the above example, income level appears to be the causative factor, which should
influence the preference for shopping centres and not the other way round. Thus, the
percentages should be computed in the direction of income level or across shopping centres.
Table 7.3 presents these percentages and suggests that the choice of shopping centres is
affected by the income level of households.
This table indicates that 67 per cent of households in the lower income level prefer shopping
centre B as against only 40 per cent of households in the higher income level. Further, 33
per cent of households in the lower income level prefer shopping centre A as compared to
60 per cent of households in the higher income level. It is apparent from Table 7.3 that
there are marked differences in the choice of shopping centres on account of differences in
household incomes.
Sometimes, the cause and effect rule leads to the conclusion that percentages might be
computed in either direction. In such cases, the researcher has to use his direction. It may
also be noted that the cause and effect rule is not always applicable. It may be advisable in
a particular problem to compute percentages in a certain direction, but the data might not
permit the researcher to do this.
Computer Processing
The use of computers in data processing has increased at a fantastic rate during the past
decade. On account of the rapid use of computers in varied spheres it is said that human
civilisation has entered the second industrial revolution. The first industrial revolution freed
man's muscles; the second is freeing his mind for more challenging pursuits. However.
processing of data by computers may not always be economically beneficial. It is, therefore
necessary to know when computers should be used. In the processing of data, computer^
may be used most efficiently if the processing operations have one or more of the following
characteristics.
I
Large volume of input: "When a large amount of data needs to be processed, computer
processing may turn out to be more economical as compared to other methods of dat;
processing.
Repetition of projects: When repetitive projects are undertaken by an organisation, the

use of the computer for processing data is advisable. In such cases too, computer processing
will generally be cheaper than its use in individual projects.
Desired and necessary greater speed in processing: When information is needed in

alimited time, manual processing of data should be given up in favour of computer processing
because of its unbeatable speed:
Desired and necessary greater accuracy: There are likely to be mistakes in data
processing by manual methods. Computer processing will be more accurate provided
sufficient care has been exercised in planning the task.
Processing complexities that require electronic help: When a number of interacting

variables are involved in the data, the computer will be most appropriate. Certain analytical
tools such as linear programming, business simulation, factor analysis and discriminant
analysis generally require the use of a computer.
The figure given below shows the average cost relationships of the different methods of
data processing.
card Average cost,
Average cost in rupees Average cost, manual
-
Average cost,
machine-assisted
\ "C" Average
cost, punched
computer
Fig. 7.1
The diagram shows that at point A there is a break-even cost position between manual
and computer processing of A' volume of data. Beyond this volume, if data is to be
processed, the use of the computer will be more economical. Likewise, points B and C
are other break-even points.
Point B shows the break-even cost position between machine-assisted manual processing
and computer processing. Thus, it may be more economical to use the machine-assisted
manual method of data processing when the volume of data to be processed is up to B.
Marketing Research
The use of computers for data up to B volume will be costlier. Beyond this point, the
computer is more economical.
However, when the volume of data is beyond B but up to C, we find that another method
of data processing, namely, that of punched cards turns out to be more econimical than
computer processing. In other words, the computer should be used when the volume of
data to be processed is quite large, in this case beyond C'.
It may be noted that these average cost curves are not stable and they frequently change
on account of changes in clerical labour rates and the cost of clerical office supplies. Over
the years, there have been significant increases in clerical labour rates and the cost of
clerical office supplies.
Such increases have shifted the manual methods cost curve upward. Side by side, there
have been innovations in computer hardware. Such innovations have shifted the computer
cost curve downward.
These shifts in the cost curves have, on the whole, made computer processing more ,
attractive for lower volumes of data processing. In general, the suitability of methods can j
be judged on the basis of the volume of data processing.
7.6 SUMMARY _________________________________________________

It may be pointed out that data processing is generally not given sufficient and advanced
attention in research investigations. This has resulted in the poor quality of data and poor
interpretation. It is desirable to have careful planning regarding how data processing will
be done, preferably at an early stage. In fact, the pre-testing of a questionnaire can be ven
helpful in determining the suitable code categories for the different responses. Designing of
data forms is equally important. s, j
'•:: }|
It may be emphasised that data processing can turn out to be both time consuming and
expensive if sufficient and timely attention is not given to its various aspects. Above all,
sound common sense coupled with experience is needed in ensuring the processing of
data accurately.
7.7 KEYWORDS
Validate: make valid
Ambiguous: double meaning

„' Monotony: lacking in variety
Drudgery: uncertain
*"*•,*
7.7 SELF-ASSESSMENT QUESTIONS

Ql. Explain data processing in your own words.
Q2. What are the different steps involved in data processing?
Q3. What is coding?
Q4. Distinguish between one way tabulation and a cross tabulation. Support your answers
with your own examples.
Q5. Prepare a two way tabulation using
^ a) Income level and ownership of a microwave oven.

•j
*' b) Use of a newly introduced cell phone from Nokia.
Marketing Research
8.1 INTRODUCTION
In the preceding units, we have discussed almost all aspects that a marketing researcher
needs to know in order to carry out his assignment successfully. At this stage, it may be
mentioned that the analysis of data does not provide the answers to research questions.
This suggests that something beyond analysis is necessary. Before the marketing researcher
prepares his report on the research done by him, he has to draw specific conclusions from
the data analysed earlier. This brings us to the interpretation of data, and as to how the
data is to be analysed and interpreted. This is very important as we have to calculate the
mean, mode and the median of the frequencies. Let us discuss these in detail.
8.2 NATURE OF INTERPRETATION _________________________________
Interpretation means bringing out the meaning of the data. Interpretation requires both,
quantitative and logical skills. Logical thought process uses two types of reasoning; inductive
and deductive reasoning.
Inductive reasoning leads from a particular instance to a general principle while deductive
reasoning moves from general principle to the specific case.
A classic example of deductive logic.
"All men are mortal" (major premise)
" Socrates is a man" (observed fact)
Therefore, "Socrates is mortal" (inferred conclusion)
Data analysis and interpretation are closely interrelated. Improper analysis will lead to '
incorrect interpretation. For a logical concrete and helpful interpretation the following things
must be kept in mind.
1. The interpretation should be honest and undistorted.

2. The objective and fundamental principle should always be dealt first.
3. Small samples have their limitations and hence cannot be generalised for a large
population.
4. Infrequent answers should not go unnoticed.
5. Evidence should be recorded without bias.

Unit 8 Data Analysis and Interpretation
6. Cause and effects, opinions and facts should not be confused or misplaced with each
other.
7. Averages are merely tendencies, not generalisations.
Data can be interpreted into a logical form using statistical techniques. Statistical analysis is
the refinement and manipulation of data that prepares them for the application of logical
inference.
Interpretation means explanation or finding out the meaning. It involves drawing inferences
from the analysis of data. Interpretation and analysis are closely interlinked. Analysis of
data often includes a simultaneous interpretation of the results.
Interpreting data in proper perspective is very important. In order to do so, the researcher
should possess a high degree of skill and exercise the utmost care and objectivity.
To begin with, the researcher must ensure that the data collected is reliable and adequate
for drawing interferences. If the data suffers from inadequacies, then even the best of the
methods used in analysis and interpretation will be rendered useless.
Any research that is based on sampling is naturally subject to a sampling error. In a sample
survey, there is a general tendency on the part of the researcher to jump to conclusions or
generalisations.
Finally, it must be remembered that the data is analysed related to a single point of time in
the past. As such, the situation might have undergone a change and one must not be rigid
in adhering to the same results where the situation is a dynamic one.
These are some of the major flaws that might occur in the interpretation of data. The
problems specified above are merely illustrative. The marketing researcher should take
care to avoid such mistakes otherwise he would reach wrong conclusions. Having
emphasised the need for proper interpretation of data, we now turn to report writing. The
rest of this chapter provides broad guidelines; for the preparation of research reports.
8.3 OVERVIEW OF DATA ANALYSIS PROCESS
Assemble the data

1 Bringing data into
order
(Arrays, tabulations, establishing categories, percentage)
A I
159
Marketing Research
Summarise the data

(Measure central tendencies, Dispersion, Graphical presentations)
Select appropriate analysis method
Examine difference Analyse experimental Investigate

Data Associations
Source - "Luck and Rubin"
Tabulation: The arrangement of data items in a numerical sequence, in either ascending

or descending order is called an array. Tabulation can take two forms namely a) simple or
b) cross tabulation.
A simple tabulation shows the distribution of one variable at a time whereas in a 1

cross tabulation the relationship between two or more variables can be studied
simultaneously, f
Simple tabulation for example,
Number of respondents reading magazine

Magazine Number of respondents % of respondents
A 10 10%
B 15 15%
C 25 25%
D 45 45%
E 5 5%
Base= 100 Respondents
Bivariate (two variables) tabulation

Household size
Consumption of 1-2 3-4 > 5 27
>5 15 25 23 11
2-5 <2 10 15 5
5 12
7 13
8.4 STATISTICS ASSOCIATED WITH FREQUENCY DISTRIBUTION

After the collected data have been processed, it is necessary that these data are analysed.
As there are several statistical techniques available to do this, the researcher has to decide
which of them he will use. In fact, a decision in this respect is called for even before the
data collection has begun so that those techniques can be used properly.
Measures of Central Tendency
I The principal measures of central tendency are the arithmetic mean, the median and the
*• mode.
The arithmetic mean should be used in case of intervalled or ratio-scaled data. It is obtained
by adding all the observations and dividing the sum by the number of observations. In case
of afrequency distribution, the arithmetic mean is obtained by the following steps:
a) the mid-point of each class interval or category is multiplied by the number of

„ observations (called frequencies) in that class,
&?'«
b) the resultant values are summed up, and the total thus obtained is divided by the total
number of observations.
Symbolically,
n
where
X = the sample mean
fi= the frequency of the i* class
xi = the mid-point of the i* class
h = the number of classes
n = the total number of observations in the sample
For a population mean, the formula is:

Marketing Research
N .
where,
the Greek letter n = the mean of the population
N = the total number of observations in a population
The mid-point is a good approximation of the true mean of the class. This is based on the
assumption that the values are distributed fairly evenly throughout the interval. When a
large number of frequencies occur, this assumption is usually acceptable.
A short-cut method by taking arbitrary mean is followed. The formula for calculation of
the arithmetic mean by the short-cut method is as given below:
X=A+- xC
n
where
A = arbitrary or assumed mean
fi = the frequency of the ith class
di = deviation from the arbitrary or assumed mean

C = common factor (or size of the class interval)
Some of the important characteristics of the arithmetic mean are:
1. The sum of the deviations of the individual items from the arithmetic mean is
always zero.
I
This means ^ (X - X) = 0, where X is the value of an item and X is the arithmetic
average.
Since the sum of the deviations in the positive direction is equal to the sum of the
deviations in the negative direction, the arithmetic average is regarded as a measure
ofcentrality.
2. The sum of the squared deviations of the individual items from the arithmetic mean is
always minimum. In other words, the sum of the squared deviations taken from any
value other than the arithmetic mean will be higher.
'':
3. As the arithmetic mean is based on all the items in a series, a change in the value of
any item will lead to a change in the value of the arithmetic mean.
4. In the case of a highly skewed distribution, the arithmetic mean may get distorted on
account of a few items with extreme values. In such a case, it may cease to be the
representative characteristic of the distribution.
The median is the measure of the central item when all the items in a series are arranged
either in ascending or descending order of magnitude. Thus, in an ungrouped frequency
distribution, if the n values are arranged in ascending or descending order or magnitude,
the median is the middle value if n is odd. When n is even, the median is the mean of the
two middle values. For a grouped series, the median is calculated by linear interpolation
with the help of the following formula:
M - t 2 + Â(M - c)
where,
M= the median
£ i = the lower limit of the class in which the median lies
^2 = the upper limit of the class in which the median lies
f = the frequency of the class in which the median lies
m = the middle item or n/2
c= the cumulative frequency of the class preceding the one in which the median lies.
Some of the important characteristics of the median are:

Marketing Research
1. Unlike the arithmetic mean, the median can be computed from open-ended
distributions. This is because it is located in the median class interval which would not
be an open-end class.
2. The median can also be determined graphically whereas the arithmetic mean cannot
be ascertained in this manner.
3. As it is not influenced by the extreme values, it is preferred in case of a distribution

having extreme values.
4. In case of the qualitative data where items are not counted or measured but are
scored or ranked, it is the most appropriate measure of central tendency.
The mode is another measure of a location of a frequency distribution. It is the value at the
point around which the items are most heavily concentrated. In case of a grouped series,
it is determined by the following formula:
Mode = L + X1
(f, - f 0 ) - (f t - f 2 )
where
^ i = the lower value of the class in which the mode lies f j=
the frequency of the class in which the mode lies ffl= the
frequency of the class preceding the modal class f 2= the
frequency of the class succeeding the modal class i= the class
interval of the modal class Some of the important
characteristics of the mode are:
• It can be applied to both qualitative and quantitative distribution.
• It remains unaffected by the extreme values in the distribution.
• It can be ascertained in an open-ended distribution.
The most commonly used statistics for measure of location are the mean, mode and median.
For variability is range, standard deviation and the coefficient of variation.
The measures of location i.e. the measure of central tendency tend to describe the
centre of distribution. If the entire sample is changed by adding a fixed constant to each
observation, then the mean, mode, and median change by the same fixed amount.
Mean: The mean or average value is most commonly used central tendency. It is used to
estimate the mean when the data have been collected using an interval or ratio scale. The
data should display some central tendency, with most of the responses distributed around
the mean.
x =
X= arithmetic mean
x = midpoint of each class
f= frequencies in each class
Y; = sum of the values
Mode: is the value that occurs most frequently. It represents the highest peak of distribution.
M =L (i)
o
mo
M =mode
0
L, = Lower limit of the modal class

Mo
d, = difference between the frequency of the modal class and the frequency of the class
immediately preceding it in distribution.
d2 = difference between the frequency of the modal class and the frequency of the class
followingit.
(i) = size of the interval of the modal class.
Median: Median is a measure of central tendency, which tends to be the average, most
typical of all values in distribution, as it is the item which lies precisely halfway between the
highest andjowes^value. It is essefltialjoârrange the data into ascending or descending
order before selecting the mectianâluil^ ——
165
Marketing Research
Lmd + (N/2)-CF
md = (i)
find
md = Median
Lmd = lower limit of median class
N = Total number of frequencies in the distribution
CF = cumulative frequencies for the class immediately below the class containing median
find = frequency of the median class

i= size of the interval of the median class
Measures of variability : A statistics that indicates the distribution's dispersion or the

spread of observations are as follows.
1. Range : is the simplest measure of dispersion. It is the difference between the

largest and smallest value of a distribution.
Range = X largest - X smallest
2. Variance : The mean squared deviation of all the values from the mean. 3
. Standard deviation : is the square root of the variance.
6 = VN -i
§ = Standard deviation
x = value of observations u
= mean of the observations N =
Total number of observations
The standard deviation of a sample is calculated as
6= -i
x = value of observations
viS
where
X = mean of the sample
n=total number of observations in sample
For grouped data the formula is

£fx* - (£fxf
6 = VN-I
x=midpoint of class
f=class frequency
Coefficient of variation: It is the ratio of standard deviation to the mean expressed as a

percentage. The coefficient of variation is useful if the variable is measured on a ratio scale.
S
CV = ---------(100)
X
CV = coefficient of variation
S = Standard deviation of the sample
X = Mean of the sample
8.5 MEASURES OF SHAPE _____________________________________
Besides the measures of variability, measures of shape are also useful in understanding the
nature of distribution. The shape of a distribution is assessed by examining skewness and
Kurtosis.
Skewness: A characteristic of a distribution that assesses its symmetry about the mean.
Mean
Mode •
Median
Marketing Research
Kurtosis: It is the measure of relative peakedness or flatness of the curve defined by the
frequency distribution. The kurtosis of a normal distribution curve is zero.
Statistical analysis appropriate to the level of measurement scale
Type of Appropriate Appropriate Appropriate Test of

scale Average measure of measurement significance
dispersion of correlation
Nominal Mode None Contingency Chi-square test
Coefficient
Ordinal Median Percentile Rank Sign test
correlation
Interval Arithmetic Standard Coefficient of T-test or F-test
Mean deviation correlation
Ratio Geometric Coefficient All of the above All of the above
mean of variation
Source : Sidney Siegel, Nonparametric statistics for the behavioural science
8.6 MEASURES OF DISPERSION __________________________________

So far the discussion was confined to the measures of central tendency. Although these
measures are useful, they alone are not sufficient as they give a very inadequate description
of the sample data. This is because the measures of central tendency are not concerned
with the variability in a distribution. For this purpose, some other measures are used.
These measures are the range, the mean deviation, the standard deviation and the variance
and they indicate how the data are spread out. As the last two of these measures are
frequently used, the discussion below is confined to them. The formula for calculating the
standard deviation in case of an array of data is
8 -
N
where
CT = standard deviation
xi = the value of the ith observation

•UER
|a=themean
N=the total number of observations
The formula for the standard deviation computed from data in a frequency distribution is
N IT =
where
fi = the frequency of the ith class

xj = the mid-point of the ith class
h = the number of classes
and all the other symbols are the same as in the previous formula.
The variance is the square of the standard deviation and is calculated by the same formula
with the square-root sign removed.
When the two series given in different units of measurement are to be compared, the
standard deviation will not be suitable. In such a case, the coefficient of variation is computed
for each series. The formula for the coefficient of variation is
where
C is the coefficient of variation, CT is the standard deviation and |a is the arithmetic mean
Since, a and u are both measured in the same units, their ratio a /|u is a pure number and
does not have any unit of measurement. A comparison of the two coefficients of variation
will indicate which of the two series having di f ferent units of measurement has greater
variability.
8.7 SUMMARY _ ' _
After collection, data has to be processed. It is necessary that this data is analysed properly.
For analysis, several statistical techniques are available which we have already studied in
this unit. The researcher has to decide which of them is to used for his study. In fact, the
169
Marketing Research
decision in this respect should be thought of even before the collection of the data. Once
the objective of the study is clear, then the right techniques to be used have to be decided.
In this unit we have studied the techniques which are all related to the analysis of the data
and the measures of the central tendency where the principal is mean, mode and median.
8.8 KEY WORDS ________________________________________________

Deviations: turn aside
Coefficient: quantity placed before and multiplying an algebraic expression
Dispersion: being dispersed

Ql. What are the characteristics of arithmetic mean? Q2.
What are the characteristics of mode? Q3. What are the
characteristics of arithmetic median? Q4. Explain the steps
involved in the data analysis process. Q5. What is coefficient
of variation?
Marketing Research
9.1 INTRODUCTION __________________

Regression Analysis is an important multivariate technique, which attempts to ascertain the
relationship between a given variable and a number of other variables. It is the most widely
applied technique amongst the analytical models of association in market research, e.g. A
marketer believes that a consumers' purchase intent is affected by her/his attitudes.
Purchase intent=Function of (Attitude 1, Attitude 2 ............. n)
For the purchase of coffee could be .

Purchase intent=Function of (Taste, flavour, colour, price, pack, .......)
To know the precise nature of this relationship one needs to do the regression analysis.
I
• Regression analysis can be done with one dependent variable and one independent
variable (simple regression) or with one dependent variable and several independent
variables.(multiple regression)
• Regression assumes that independent variables take fixed values and can be measured
on any scale.
i
i
• Dependent variables must be measured on a ratio/interval scale.
• Dependent variables can be expressed as a linear function of independent variables,

•
Let us take a simple illustration of one independent variable (variable x= advertisement

expenditure) and one dependent variable (sales=Y). The linear relationship will be as
shown in the graph.
Y=Retail sales of consumer product (in thousands of unit for defined time period)
X=Advertising expenditure (in lakhs of rupees during the same time period)
N=Number of markets from where the sales data has been obtained
The relationship between the two variables will be expressed by the simple regression
equation.
«:
Unit 9 Multivariate Analysis
where
a=expected values of sales at X = 0
b = average change in sales per unit change in advertising expenditures. Also called the
coefficient of gross regression of Y on X.
Y C =Sales forecast for given value of b X
We have to find the values of a and b in this equation such that the deviations between the
yx
., original observations and calculated values of the dependent variable measured by the
following formula are at a minimum.
where •
i • "
•| Y. = value of sales in the ith market yc2
= variance yc = standard deviation of
regression
For obtaining the values of a and b in this average linear relationship between the variables,
the method of ordinary least squares is used.
Given below are the standard statistical formulae, which are used in this method. The
least squares
Covariance (y, x)
Coefficient of gross b =
Variance (x)
yx
Regression
=
Once the value of b is calculated, it can be substituted into the following formula to obtain
the least-squares estimate of the intercept.
a=Y-b X
yx
Once the values of a and b are known, we can form the regression equation
yx
Y=a + b X
c yx
Marketing Research
For a defined value of X (advertising expenditure), we can obtain the sales forecast yc.
The multiple regression using two independent variables and a dependent variable will
give an equation.
Y =a + b .z +b . x
c yx x yz z
yz
With the past data available, values of a, b .z and b . x will have to be obtained. The
yx
mathematical operations become more complex with increase in the number of independent
variables.
Points to Ponder
• Equations should be validated statistically in multiple regression analysis.

i
• If the past observations are too less than the usage of independent variable gets
restricted.
• If independent variables are highly correlated among themselves they are facing the
problem of multicollinearity. One way to overcome multicollinearity is to drop certain
variables from the model if the corresponding standard error of regression coefficient
is unduly large.
• When the variables are too many and the analysis is getting complex, stepwise
regression should be followed.
• A limitation with regression forecasting is that independent variables must be forecasted

first before forecasting the dependent variable.
Thus, regression equations can be used for problems involving prediction and forecasting.
It helps study how much of the variation in the dependent variable is explained by the set
of independent variables.
9.2 CONJOINT ANALYSIS ______________________________________

Conjoint analysis attempts to determine the relative importance consumers attach to salient |
attributes and the utilities they attach to the levels of attributes. The consumer is presented
the stimulus in the form of totality of the product and not asked to score individual attributes
separately. However, the stimuli is created by the researcher himself unlike multi dimensional
scaling where attributes to be studied previously exist.
Unit 9 Multivariate Analysis '
• the stimuli created is typically used to identify the most desirable attributes in a product
or service. Consumer preferences can be evaluated through the 'trade off' of attributes
made by them.
• applies complex Anova techniques to the preference data obtained from the
respondents. A value (utility) is calculated for each feature. Features with highest
value are considered most important to the respondent.
• is applied to categorical variables, which reflect different features of the product

under consideration.
• identifies interdependences among variables.
• measures the relative importance of various combinations of the features/attributes.
Assumption
• A consumer has to choose from multi feature products or options.
i An individual is capable of ranking options on basis of some criterion.
• Ranking involves complex trade offs i.e. to obtain certain desired quality product
attributes some other attribute will have to be compromised.
Utilities
Conjoint analysis generates utilities for each level of each attribute for every individual
respondent.
• When the utility of one attribute is added to utility of another attribute(s) the sum for
combination displays good correspondence with that combination position in the
respondent's original preference ranking.
I
• A respondent's first preference is supposed to have the highest utility, the second
preference the next largest utility and so forth. A feature what a respondent is reluctant
to forego is considered to have high utility.
« Utilities indicate the relative importance of each attribute level or feature.
t Utilities are arrived through iterative techniques, trying to fit different numbers,
using a predetermined logic to satisfy the requirements as stated below.
177
Marketing Research
a. Identify the utilities for each attribute level included in an alternative.
b. All utilities must be added for all attribute levels included in that attribute.
c. Ranking of total utility is done from highest to lowest level.
d. To check ranking is similar to ranking obtained, respondent could be asked to rank

product features in a 10 point scale.
e. Utilities based on relative preferences for combinations may not match expressed
preferences if combined utilities of two or more combinations are similar.
f. If utilities are consistent across respondents then the standard deviation of scores of
each level of attribute must be studied.
Example
Attribute Utility Relative importance
A 1.2 8
B 1.8 6
C 1.5 7
D 3.1 3
E 2.5 4
F 4.5 1
G 3.9 2
H 2.0 5
Combination Combined Relative preference
BAD (1.8+1.2+3.1+2) 8.1 6 Least preferred
H
ACEG 9.1 5
AEFG 12.1 2
BDEF 11.9 3
GFDC (3.9+4.5+3.1+1.15) 13.0 1 Highest preferred
CDEF 11.6 4
9.3 APPLICATIONS AND PROBLEMS OF CONJOINT ANALYSIS
Conjoint analysis has been used in marketing for varied uses.
Market Segmentation: Through conjoint analysis individual customers identity can be

retained and consumers with homogenous preferences can be targeted as a segment.
Competitor's vulnerability can be detected and at the same time self-cannibalization effect
of additional new products can be studied.
Estimating market share of brands: Utilities derived from conjoint analysis can be
used as input into a choice stimulator to determine the share of choices and hence share of
brands.
Optimum product design: Brand features can be varied in terms of attribute level and
corresponding utilities. The Brand features yielding the highest utility indicate the composition
of the most preferred brand. Company's product can be measured with competitor's
product and what attribute changes is likely to lead the company into can be studied.
Relative importance of attributes in the consumer choice process can be determined.
Companies can differentiate customers as' loyals', 'acceptors' and' switchers' for more
accurate brand equity calculation.
Conjoint Analysis enables companies to conduct brand swot analysis.
Problems in using conjoint analysis
1. Conjoint analysis assumes that the most important attributes are included in the study.
I
2. It assumes the addivity of utilities for combination of attributes.
3. Problem could arise with the appropriateness of the levels or features used for each
attribute in the study. Decisions could be taken which do not capture the utility •Wip
adequately.
4. Validating the model is quite difficult as it is relatively new.
5. It may be difficult to ascribe and describe attributes for certain products.
Researches have indicated that a lot of consumer and industrial goods companies are
relying on conjoint analysis for product attribute development within the given cost structure
of the product possible for the company.
1TQ
Marketing Research
9.4 FACTORANALYSIS ________________________

Factor analysis is a multivariate statistical technique that uses a class of procedures for
data reduction and summarisation. Unlike regression the entire set of interdependent
relationships are examined.
1. Uses an advance form of correlation analysis to the responses.
2. Variables that are interrelated / similar are reduced to smaller numbers of uncorrelated
factors, e.g. Variables

Factor
Input
Output
Punctual
Trustworthy
Good employee
Knowledgeable
Disciplined
Self-initiated
^
3. The model is primarily based on linear relationship through which factor loading is
derived.
4. Factor Analysis is performed on intervally scaled or continuous variables.
Three important measures

1. Variance: Factor analysis tries to best fit the factors in a scatter diagram in such a
way that explains the variance associated with the response to each statement. ,„
2. Standardisation of response
mean of total
Individual's Individual's
Standardised = actual responses
response to the statement
Score on a to the statement
statement
Standard deviation of the total responses to the statement
•* " ' .
I
3. Correlation coefficient: This analysis computes correlation coefficient associated

with the standardised scores of the responses to each pair of statement. Factors are
generally uncorrelated to each other.
Factor Analysis Model

Factor Loading is the correlation between a factor and a variable (the statement's
standardised response score). The greater the value of factor loading, the better is the fit
of the factor to the data from the concerned statement. These loadings have a lower limit
of-1.0 and an upper limit of+1.0.
Eigen value: Each factor has an eigen value. It is the measure of variance explained by
each factor. A factor's eigen value is the sum of squares of its factor loading.
Communality is the amount of variance a variable shares with all other variables included
in the analysis, e.g.
Factor Communality
1 2 3
Eigenvalue 2.23 1.32 0.60 —
Explained
Variance 0.45 0.26 0.12 —
Statements
1 0.91 0.06 0.11 0.84
2 0.89 0.14 0.24 0.87
3 0.75 0.25 0.10 0.64
4 0.20 0.86 0.18 0.81
5* 5
0.10 0.70 0.70 0.99
In the above example
Fl is a good fit on statements 1 , 2 and 3 but a poor fit on 4 and 5. This could possibly
show that statements 1-3 measure same basic aspect.
InFl eigen value = (0.91)2 + (0.89)2 + (0.75)2+(0.20)2 +(0.10)2 = 2.2371

3. Correlation coefficient: This analysis computes correlation coefficient associated

with the standardised scores of the responses to each pair of statement. Factors are
generally uncorrelated to each other.
Factor Analysis Model

Factor Loading is the correlation between a factor and a variable (the statement's
standardised response score). The greater the value of factor loading, the better is the fit
of the factor to the data from the concerned statement. These loadings have a lower limit
of-1.0 and an upper limit of+1.0.
Eigen value: Each factor has an eigen value. It is the measure of variance explained by
each factor. A factor's eigen value is the sum of squares of its factor loading.
Communality is the amount of variance a variable shares with all other variables included
in the analysis, e.g.
Factor Communality
1 2 3
Eigenvalue 2.23 1.32 0.60 —
Explained
1
Variance 0.45 0.26 0.12 —
Statements
1 0.91 0.06 0.11 0.84
2 0.89 0.14 0.24 ,
V
3 0.75 0.25 0.10 0.87
0.64
4
*»'• 0.20 0.86 0.18 0.81
5 0.10 0.70 0.70 0.99
In the above example
Fl is a good fit on statements 1,2 and 3 but a poor fit on 4 and 5. This could possibly
show that statements 1 -3 measure same basic aspect.
InFl eigenvalue = (0.91)2 + (0.89)2 + (0.75)2 + (0.20)2 + (0.10)2 = 2.2371
181
Marketing Research
Communality
Statement 5 =(0.10) 2 +(0.70) 2 +(0.70) 2 = 0.99
How many factors should be used? (Number)
Since factor analysis is designed to reduce many variables to fewer underlying factors, a
critical question is how many identified factors should be used. These are few of the
approaches to number determination.
A prior determination: The researcher knows how many factors to be expected and
thus can specify in the computer programme the number of factors to be extracted.
Based on Eigen Values: Factor with eigen values greater than 1 are retained and the
others are excluded from the model.
Based on variance percentage: In this method the cumulative percentage of variance i

extracted by the factors should reach a satisfactory level. A satisfactory level will depend I
on the problem. However, factors extracted should account for at least 60% of the variance,
Based on split half reliability: The sample is split in half and factor analysis is performed
on each half. Only factors with high correspondence of factor loadings across the two sub
samples are retained.
Based on significance test: The statistical significance of separate eigen values should
be obtained and only those factors should be retained which are significant.
9.5 MODEL, APPLICATION AND LIMITATIONS OF FACTOR ANALYSIS

1. Factor analysis can be used for market segmentation studies.
2. Determining brand attributes that influence consumer choice.
3. Understand media consumption habits of target market e.g. frozen food users watch
cable channels.
4. Identifying price sensitive consumers for pricing studies.
Limitations of factor analysis
1. How many factors should be employed for reducing the data? It is quite a
subjective decision.
Unit! Multivariate Analysis
2. What criteria should be used in establishing that number is a complex question.
3. Labelling of factors is purely intutive and subjective.

4. Sometimes, different factor analysis procedures may give different results.
Despite the limitations like all mathematical models the usefulness of factor analysis cannot
be undermined.
9.6 CLUSTERANALYSIS _______________________________________

Cluster analysis is used for market segmentation. Through this technique objects/cases are
classified into relatively homogeneous groups called clusters. Objects in a cluster are similar
to each other but tend to be dissimilar with other clusters. Cluster analysis has been used
in marketing for the following purposes.
1. Segmenting the market on the basis of benefit sought from the product.
2. Understanding buyer's behaviour.

3. Identifying new product opportunities. Brands in the same cluster tend to compete
more with each other than product falling under different clusters.
4. Selecting test markets for various marketing strategies.

5. Making data manageable by clustering and reducing individual data study.
Clusters can be made by buildup (combine entities) techniques or teardown (divide entities
to form clusters) techniques.
Variable 1 An ideal Variable 2 A practical clustering
clustering situation situation
Conducting cluster Analysis
Define Inputs
Compute distance
Select a clustering procedure

1
183
Marketing Research
Decide the numbers of clusters 5

1
Interpret the cluster profile , ,,
1 ' ^
Assess validity
1. Define inputs: Selecting the variables on which the cluster would be based is the
most important part in cluster formation. Variable could be demographics (age, sex,
income, geographic location), psychographic (activities, interest, opinions) ormulti
factors. Companies must identify their own product capabilities, financial strength,
distribution strength and who are they going to serve.
e.g. washing powder
Rank Respondent 1 Respondent 2

Foams well 6 86
Cleans well 5
2. Computing distance: Similarity is measured in terms of distances between pairs of

objects. Lesser the distance greater is the similarity. The two methods of computing
the distance are as given below.
a. City block metric: In this method the difference in ratings for each attribute1!1; I
added.
Rl R2 Difference
Foams well 6 8 2
Cleans well 5 6 1
The distance = 3
b. Euclidean
distance
Add the (difference)2
Distance (2) 2 + (l) 2
In this cased 5 4.
2.2
3. Select a clustering procedure
Clustering
Hierarchical (New Non Hierarchical or

clusters are based K means clustering
on previously (Clusters are created
formed clusters) new at each stage)
Agglomerative Divisive
(groups objects into (Dividing into Sequential Parallel
bigger and bigger smaller and Threshold Threshold
clusters) smaller
clusters) Optimising
-Linkage Partitioning
-Variance -
Centroid method
Hierarchical methods
Linkage methods
K - mean / nonhierarchical clustering
a. Segmental threshold method : A nonhierarchical clustering procedure
determines a cluster centre and then groups all objects within a prespecified
threshold value from the centre.
b. Parallel threshold method: Several cluster centres are specified at once and
all objects that are within a prespecified threshold value from the centre are
grouped together.
c. Optimising partitioning methods: This method allows for later reassignment

of objects to clusters to optimise an overall criterion.
4. Decide the number of clusters: There is no thumb rule to decide on the number of
dusters. However, some methods used for deciding numbers are as follows:
a. Theoretical, or practical considerations may suggest some number.
b. The distances at which clusters are combined can be used as a criteria.
185
Marketing Research
c. Relative size of clusters could be meaningful. There is no point in having a one-

object cluster.
d. In non-hierarchical clustering, the ratio of total within-group variance to between

group variance can be plotted against the number of clusters. The point at which
a sharp bend occurs usually indicates an appropriate number.
5. Interpreting the cluster profile: This involves examining the cluster centroid. The j
centroid represents the mean value of the objects contained in the cluster on each of j
the variable.
Assessing reliability and validity
Formal procedures for assessing reliability and validity of clustering solutions are complex j
and not fully defensible. Hence, validity of results could be checked by, using different ]
methods of clustering and comparing results, splitting clusters into two sub samples from
centroid and performing the cluster again or deleting variables randomly to check result of j
reduced set of variables.
Some factors to be considered while using cluster analysis technique are as follows:
• Limited number of cluster can be considered by a company due to limitation of market]

potential within a cluster and the difference between the clusters may not be I
marked.
• Characteristics of clusters may change over a period of time owing to socioecon

changes.
• Clusters are not water tight compartments. Flows between clusters may take p
e.g. People liking product X in cluster A and may also like product Y.
• Evaluating the quality of cluster is difficult and so is the selection of clustering criteria
• Clustering is based on the assumption of similarity between objects and there is j|

theoretical justification and basis for structuring objects into clusters.
• Clusters are used for further development of marketing strategies especially in c

of product development, advertising and pricing strategies.
It can thus be concluded that cluster analysis is a scientific method that helps to s
number of attitude and behavioural variables for market segmentation.
9.7 MULTI-DIMENSIONAL SCALING (MDS)

Multidimensional scaling is a statistical technique that represents perceptions and preferences
of respondents as points in a geometric space. It is a special representation of relationship.
MDS involves two process namely.

• Identification of attributes/dimension on which consumers perceive or evaluate objects.
• The arrangements of objects with respect to dimension.
Multi-dimensional scaling
\ Non
Attribute data attribute data
Similarity Preference
MDS does not explain the perception about brands but represents psychological relation
among stimuli. The psychological dissimilarity is represented as geometric distance. The
advantage of MDS lies in the fact that both metric and non-metric techniques can be used.
When the input data is interval-scaled or ratio-scaled, it is called as metric MDS. When
input data is in nominal or ordinal form it is called non-metric MDS. Whichever method of
MDS one selects, it does not change the relative distance of the points.
Data input could vary.
i. Single mode, two-way data

1 2 3
Brands E{
Here, n (n-l)/2 pairs are evaluated
B,
2. Two mode, two- way data

A3 Attribute
Brands B,
B2
B,
Marketing Research
3. Single mode, two-way data: When research data is converted into geometric
distance and arranged in a space, the closer the brands means greater similarity and
vice versa. In this picture there are 10 brands. A3 and A4 are similar to each other and
different from others.
i
4. Two mode, two-way data: Here brands are located on attribute vectors. Products
can be compared on both 'objective' and 'perceived' attributes. Buyers perceptions
are a sum total of company's product, competitor's products, advertising, after sales
services etc. In term a composite whole other than just the company's product. Hence,
the 'perceived' and 'objective' dimension may not be necessarily the same.
e.g. 50 MBA students were asked to compare six leading companies where they would
like to be employed on various criterion like salary structure and growth opportunity,
reputation, nature of work, posting location, office hours etc. The companies were to be
marked on similar/dissimilar attributes. The results were gathered and analysed on multi-
dimensional scaling software. The two most important parameters that emerged were
salary and growth opportunity.
Growth opportunity
High Salary
Company offering higher salary and high growth opportunities for employment were
preferred.
Uses of multi-dimensional scaling

c)
• MDS has been used to measure image of the company. This helps in knowing
perceptions of customers and non-customers and the difference with the perceptions
of the company itself.
• Market Segmentation: Companies can identify groups with similar perceptions

and preferences and modify their product offerings to an' ideal' requirement of tk
customer.
• Vendor evaluation: has to be done to select the best supplier for various compan
purchases. Vendors can be evaluated for price, reliability, credit, technical suppon
and timely deliveries.
• New product development: Identification of spatial gaps in the product percepfiffl'

maps can help develop new products for the target consumers.
188
» Assessment of advertisement effectiveness : Assessment of advertisement is

possible through perception maps. It can be adjudged whether the brand message
has been rightly received and brand positioning achieved.
• Test marketing: Cities with similar demographic characteristics could be identified

to assume national response and test marketing in these select cities can be done.
« Product life cycle: Current perceptions about a product can be studied in relevant
segment and products which are registering declining sales curve can be repositioned
with required alterations.
• Advertising Media Selection : Target audience could be located at their ideal

point and different media could be identified in their similarity space to choose the
media for advertising.
• Pricing analysis : The impact of pricing can be studied through comparison of

spatial maps with and without pricing information.
There exist certain limitations in the usage of this techniques. They are as follows:
•ft
a) Conceptual problems: Accurately defining similarity and preference and communicating
exactly to respondents seems non-achievable goal.
b) Computer programmes aid MDS. Different programmes used for analysis may have
different distance functions.
c) The data collection process is subject to bias as respondents preferences are a

subjective issue.
Despite its limitation MDS is a popular, technique used by students and marketers alike.
>& Activity A;
Study the products and consumers of a company you are familiar with and develop a
cluster of customers depending on their different requirements.
189
Marketing Research
9.8 SUMMARY
In this unit, we have studied the different types of statistical techniques which are been
used for the analysis of data. Now the question arises into our minds, why statistics is
used? Collection of data or designing of questionnaire is a part of survey, but at the same
time when the technique that will suit our study and the objective of the study has to be
fulfilled, it might be the conjoint analysis or it might be multidimensional scale etc. but to fit
the exact variable and dependent variable, we need to know the correlation analysis of the
data collected. Statistical techniques play a very important role in Market research.
9.9 KEYWORDS ______________________ , _________

Regression: move backwards
Stimuli: animate, excite, arouse
Iterative: repeat Intuitive:
possessing
9.10 SELF-ASSESSMENT QUESTIONS ____________________

Ql. The technique which helps to express the dependent variable in terms of the
independent variable is:
a) Correlation analysis b) Regression analysis
c) Multi-dimensional scaling d) Cluster analysis
Q2. The technique which aptly identifies target segment of homogenous consumers is
a) Regression analysis c) Cluster analysis
b) Factor analysis d) Conjoint analysis
Q3. Consumer preferences can be evaluated through 'trade off' of attributes made I);
them in:
a) Correlation analysis c) Multi-dimensional scaling
b) Conjoint analysis d) Cluster analysis

Marketing Research
10.1 INTRODUCTION
The research process consists of a series of steps that cover the research project from its
conception to final recommendation stage. The various steps in the research procedure
are as follows:
• Determination of the purpose of the study. Problem and opportunities whether overt/1
latent/proactive have to be determined to set the research process rolling.
• Establish research objectives.
• Evaluate the (cost v/s benefit ratio) value of the research information.
• Design the research methodology. Thi s includes primary and secondary data collection. ]
• Implementation of the chosen research approach.
The market research process involves a number of inter-related activities which overlap
and do not rigidly follow a particular sequence. A researcher is often required to think a
few steps ahead. For example, if a researcher has formulated a research problem and is
considering the sampling plan, he is supposed to consider the type of data to be collected
as also the detailed tabulation. This is because the various steps are inter-woven into each
other and each step will have some influence over the following step.
In market research, even though our focus is on one particular step, other inter-related
steps or operations are also being looked into simultaneously. As we complete one activity;
or operation, our focus naturally shifts from it to the subsequent one, i.e. the focus is not:
concentrated exclusively on one single activity or operation at any particular point of time.;
Let us now describe the major steps involved in a market research project.
10.2 RESEARCH DESIGN ___________________________________

Research design is a framework or blueprint for conducting the market research pnî It
specifies the details of the procedures necessary for obtaining the information neededio \
structure and/or solve market research problems. Aresearch design is formulated priors? the
commencement of the project. A good design ensures that required data is gathcres
accurately, timely and economically.
10.3 FORMULATING THE RESEARCH PROBLEM _____________

The first step in research is formulating a research problem. It is the most important si in
194 applied research as poorly defined problems will not yield useful results. It is rightlysf
Unit 10 Research Process and Design
that "a problem well defined is half-solved". Poorly defined problems cause confusion and
do not allow the researcher to develop a good research design.
In order to identify the research problem, three categories of symptomatic situations, namely,
overt difficulties, latent difficulties and unnoticed opportunities should be studied. Overt
difficulties are those which are quite apparent and which manifest themselves. For example,
if a firm has been witnessing a decline in its sales for some time, this could be called an
overt difficulty. Latent difficulties, on the other hand, are those which are not so apparent
and which, if not checked, would soon become evident. For example, declining sales may,
in due course, demoralise the sales staff. Unnoticed opportunities indicate the potential for
growth in a certain area of marketing. Such opportunities are not clearly seen and some
efforts are required to explore them.
It is difficult to lay down any concise prescription for recognising problems. A person with
an inquisitive nature and the necessary background would recognise a problem or an
opportunity in less time than another who lacks these qualities. Once the researcher has
identified two or more problems or opportunities, the next question he should be concerned
withis-which of the problems is to be selected? This is necessary as he will not be in a
position to take up all the problems on account of limited finances and time constraints. In
such a case he has to determine priorities, carefully examining their importance to his
organisation. Choosing a relatively less important problem would amount to wasting limited
resources. He should look into the value and cost aspects and then select that problem
which gives the maximum net value of research.
After a problem has been chosen, the next task is to formulate it precisely. This too needs
a good deal of care on the part of market researchers. Formulation implies a clear statement
or definition of the problem.
A complete problem definition must specify each of the following:
» Unitofanalysis
i Time and space boundaries
• Characteristics of interest (both the 'results' that are of concern to management and
the 'variables' that are to be tested for their relationship to the results)
» Specific environmental conditions
Taken together these four aspects identify the who, when, where, and what of the research.
These are briefly explained as below.
195
Marketing Research
Unit of analysis
The individuals or objects whose characteristics are to be measured are called the units of
analysis. The units always identify the objects to be studied. It is necessary that the universe
is well defined. Consider, for example, the statement "Women's dress buyers in Big bazar
stores on January 30, 2007". This specifies a particular universe, provided that clear
definitions are given for 'Women's dress buyers', and 'Big Bazar stores'. Consider another
universe "Women living in the Mumbai metropolitan area who are shopping for one or
more dresses in January 2007". The difference in the two statements is that whereas the
units of the universe are 'buyers' in the former, in the latter they are' shoppers'. Also, note
another difference between the two universes. In the first case, the universe indicates
'Buyers of women dresses', implying that the buyer may be either male or female. But in
the second case, only women comprise the universe.
Time and space boundaries
As regards time and space boundaries, we find that the two universes are again different.
In the first instance, a precise date, viz. January 30,2007 is given while in the second
instance the entire month of January is given. Similarly, the two universes are different in
terms of space—the 'buyers' universe specifies stores located in Mumbai while the
' shoppers' universe specifies the Mumbai Metropolitan area which should be a larger
territory than the former. A more subtle difference between the two universes can also be
seen. The 'buyers' universe specifies that buying takes place in stores located in Mumbai.
The 'shoppers' universe does not specify as to where shopping takes place. It says that
women shoppers living in the Mumbai Metropolitan area in January 2007 are shopping
They may be shopping outside Mumbai as well. Thus, in the second case the area in which
shopping occurs is unlimited. |
Hence, marketing managers run the risk of making the right decision at wrong time, *
opportunities are transient. ^
I
In view of this, it is vitally important that the marketing manager and researcher decide
upon the suitable time reference for the decision.
Characteristics of interest
This aspect identifies the focus of the problem. In our earlier example, the characteristic
of interest can be style and colour preferences, buying behaviour, personality traits, en
Again, the researcher may be interested in only one characteristic.
It is necessary that the problem definition specify one or more characteristics to be measured
and the fact that the nature of relationships amongst them is to be determined.
Thus, we may like to know more specifically as to what dresses are liked by educated
women or those who are employed. Is there any preference for store location amongst the
members of the universe on account of their income? This and similar other questions will
lead us to focus attention on the nature of relationships amongst the various characteristics.
Environmental conditions
This aspect indicates the uniqueness or generality of the problem. For example, if the
management is interested in knowing how the units respond to price changes, then the
problem definition should specify the prices to be researched.
The management is sometimes interested in knowing the behaviour of certain types of

firms under specific economic conditions. In such cases, the problem definition must spell
out those conditions precisely.
In other words, the problem definition must specify the environment for which the company
wants research results. It may also spell out the possibilities of changes as well as the
direction of change in the environment so that the results of the research study do not
become irrelevant.
It may be emphasised that the problem definition in market, research is a step towards
identification and structuring of the management's question. The most important objective
of problem definition, however, should be to answer the right question.
Hypothesis development
Before we pass on to the next stage, it is worthwhile to briefly mention the development of
hypotheses. A hypothesis is a proposition which the researcher wants to verify. Often
there may be several competing hypotheses, either specified or implied. If, before
undertaking the research, the researcher finds that all hypotheses are true, then there is no
need whatsoever to undertake research. One objective of research is to select among the
possible hypotheses and to test them empirically with the help of statistical tools in order to
ascertain whether they are true or false.
While the formulation and testing of hypotheses are important in research, it is not necessary
that every market research study must have a hypothesis. In some studies we are only
interested in knowing factual information and hence there is no need for formulating a
hypothesis.
197
Marketing Research
To sum up, a careful formulation of the research problem would be helpful in providing a
sense of direction to the research staff. As it specifies the precise scope of the problem, it
makes research both meaningful and economical. Further, problem formulation, by setting
out assumptions, would avoid any confusion to the reader. This also gives an idea of the
environment in which the research is to be done, so that focus on the problem is not lost.
Finally, problem formulation would also indicate the limitations of research itself so that
one can see it in a proper perspective.
10.4 FEATURES OF A GOOD RESEARCH DESIGN __________________

A research design specifies the methods and procedures for conducting a particular study.
The researcher should specify the approach he intends to use with respect to the proposed
study. Broadly speaking, research designs can be grouped into three categories—
• exploratory research
• descriptive research
• causal research
An exploratory research focuses on the discovery of ideas and is generally based on

secondary data. It is preliminary investigation which does not have a rigid design. This is
because a researcher engaged in an exploratory study may have to change his focus as a
result of new ideas and relationship among the variables.
A descriptive study is undertaken when the researcher wants to know the characteristic!
of certain groups such as age, sex, educational level, income, occupation, etc. In contrast
to exploratory studies, descriptive studies are well-structured.
A causal research is undertaken when the researcher is interested in knowing the cause
and effect relationship between two or more variables. Such studies are based on reasoning
along well tested lines.
It may be emphasised that the main criterion of a good research design is that it must
answer the questions posed earlier. Further, the researcher should select that researd
design which is appropriate in achieving the objectives of the study. A point wort
emphasising here is that there is no one 'best' research design. There are several alternate
methods for solving a particular problem. Therefore, the research should not be deferoi
in the quest for the 'ideal' research design. It is through experience that one is ablet
select the most appropriate research design.
A good research design provides for the minimum details required for
pi arming. It includes
l the following:
• Statement of study's objective
• Statement of data inputs or causal data on the basis of which solution is to be reached
» The analytical framework on which inputs will be treated or calculated
• The details for costing and implementation of the work
» Form and scope of the final research work
10.5 CLASSIFICATION OF RESEARCH DESIGN________________________
All research approaches can be classified into one of the three general categories of research:
• Exploratory
• Descriptive
• Causal
Each of these categories differs significantly in terms of research purpose, questions,

precision of hypothesis and data collection techniques used.
Exploratory research
Exploratory Research is used when one is seeking insights into the general nature of the
problem. There is less previous knowledge to build on. The sample research process is
flexible and unstructured. The sample is small and is non-representative of the class. The
hypothesis is tentative and a partial answer to the research question. It is generally followed
further by conclusive research. The process of gathering information is through:
• personal interviews
• industry expert views
i existing secondary data
• qualitative research
• pilotsurveys
Marketing Research
Exploratory research help set the priorities among research questions and learning the
practical problems of carrying out the research they.
• define the problem more precisely
• identify alternative course of action
• develop hypothesis and an approach to the problem
Thus, this low cost and low risk form of research can yield rich dividends.
Descriptive research
Descriptive research is a type of conclusive research, which describes some aspect of

market characteristics or function.
e.g. • The proportion of adult population reading Economic Times.

• Reader's evaluation of Economic Times versus other Financial dailies.
• • The socio economic and demographic characteristics of the readers of the

newspapers.
Such research is marked by prior formulation of specific hypothesis. Hypothesis is a tentati\ e ,

answer to a question. It is preplanned with a structured design and large representative j
sample size. The descriptive designs require a clear specification of the who, what, when, j
where, why and way of the research. The analysis is quantitative and is used as inputs for |
decision-making.
The methods used for descriptive research include,

• secondary data
• surveys
• panels
Symptoms of the problem should be distinguished from the problem as the underlying
cause may be elsewhere e.g. decrease in sales could be because of pricing, promotion x
distribution problems.
TOO
'nit 10 Research Process and Design
Secondary data
All existing sources of data must be tapped to gain information on the subject.
Surveys
Survey is the most commonly used method. It is carried with the help of questionnaires
and observation. Alarge sample size is interviewed, whereby adequate planning is required.
Panel design
In panel design method, a fixed sample of respondents provides data on a continuous

basis as against survey information, which is just once. e.g. TV channel ratings.
Causal research / experimental research design
Itis conducted to determine cause and effect relationships between one or more independent
variables. The other mediating variables are controlled. They are similar like laboratory
experiments where what goes in, what changes are made and what results are known.
Determining sources of data
The next step is to determine the sources of data to be used. The market researcher has to
decide whether he has to collect primary data or depend exclusively on secondary data.
Sometimes, the research study is based on both secondary and primary data.
When a study is to be based on secondary data, whether partly or fully, it is necessary to

note that the data are quite suitable for the objectives spelt out by the study. It is also
advisable to evaluate secondary data in details to avoid possible sources of error.
To begin with, one should be familiar with the authentic sources of relevant data, their
periodicity, the agency which is publishing or having such data, the concepts used in
compilation and their limitations, if any. A sincere effort must be made to look into the
existing data with a view to examining their suitability for the research. It is only when such
secondary data are unavailable, inadequate, or unreliable, that a researcher should decide
on collecting fresh data.
Designing data collection forms
Once the decision in favour of collection of primary data is taken, one has to decide the
mode of collection. The two methods available are as follows:
201
Marketing Research
II • Observational method
• Survey method
Observation
This method suggests that data are collected through one's observation. If the researcher
is a keen observer, with integrity he would be in a position to observe and record data
faithfully and accurately. While the observational method may be suitable in case of some
studies, several things of interest such as attitudes, opinions, motivations and other intangible
states of mind cannot be observed. Another aspect of this method is that it is nonreactive
as data are collected unobtrusively without the direct participation of the respondent. This
is a major advantage as the behaviour can be recorded without relying on reports from the
respondents.
Surveys
I i In market research, field surveys are commonly used to collect primary data from the
respondents. Surveys can be
• Personal
• telephonic
• mail
• diary
Of these, personal and mail surveys are more frequently used in our country. A choice b
to be made regarding the type of survey for collecting data. There are certain advantage*
and limitations of each type of survey. Broadly speaking, telephonic survey is suitable
when very limited information is sought in a short period of time. Moreover, such information
should be readily available with the respondents. In contrast, surveys based on personal
interviews are suitable when detailed information is to be collected. Sometimes a combination
of two or more methods could also be used.
It is a common practice to use structured questionnaires prepared in advance, to elicit^

necessary information from the respondents. In case the enumerators are to fillupll
questionnaires, the survey is a personal one. It is a mail survey if the information is s by
sending the questionnaire by post. Whether it is a personal or a mail survey, it is ne to design
a suitable questionnaire, conduct a pilot survey and undertake a pre-testing^ the
questionnaire. The pre-testing will enable the researcher to realise the shortcoming! his
questionnaire. In the light of this 'reaction' of the respondents, coupled with the f
observation of the researcher, the questionnaire should be modified.
Determining sampling design and sampling size •
Another aspect which forms a part of research process is the sampling plant. When the
Market researcher has decided to carry out a field survey, he has to decide whether it is to
be a census or sample survey. In almost all cases, a sample survey is undertaken on
account of its overwhelming advantages over a census survey.
When a decision in favour of a sample survey has been taken, it is necessary to have a
clear definition of the population from which the sample is to be drawn, before deciding on
the type of sample design to be used. First, a broad choice is to be made between probability
sampling and non-probability sampling. The researcher then selects a specific type of
sample design from a number of sample designs. The type of sample design chosen will
depend on its suitability and also the availability of the requisite sample frame.
As regards sample size, there are two basic approaches—the adhoc or practical approach
and the statistical approach. Although the former is widely used in Market research, it
should be the endeavour of the researcher to follow the statistical approach which is
based on well-defined principles. The size of sample will depend on the degree of precision
required as also on the cost considerations.
The object of sampling is to choose a sample which will faithfully reproduce the
characteristics of the population or universe. In practice, however, this objective is never
completely attained on account of the occurrence of two types of errors, due to bias in the
selection and errors due to sampling. It is desirable to minimise these errors and to consult
an experienced statistician on sampling.
Organising and conducting the field survey
Having prepared the questionnaires and selected the sample design and size of sample,
the next step is to organise and conduct the field survey. Two important aspects should be
looked into i.e.
a) interviewing
b) supervision of field work
The task of interviewing seems to be simple but, in reality, it is one of the most difficult
tasks in market research. This is because respondents are generally hesitant in giving
information unless approached with tact, initiative and intelligence. Supervision of field
work is equally important to ensure timely and proper completion of the field survey.
Neglecting these aspects would result in interviewing errors, which, in turn, would undermine
the utility of the survey.
203
Marketing Research
Processing and analysing the collected data

Once the field survey is over and questionnaires have been received, the next task is to
aggregate the data in a meaningful manner. A number of tables are prepared to bring out
the main characteristics of the data. The researcher should have a well thought out framework
for processing and analysing data, and this should be done prior to the collection. It is
advisable to prepare dummy tables, as such an exercise would indicate the nature and
extent of tabulation as also the comparisons of data that can be undertaken.
In order to derive meaningful results from the statistical tables, the researcher may use one
or more of the following four steps.
a) The first step is to calculate relevant measures of central tendency as also of dispersion,
highlighting the major aspects of the data.
b) The second is to cross-tabulate the data to ascertain some useful relationships.
c) The third is to calculate the correlation coefficient and undertake a regression analysis
between variables.
d) The fourth is to undertake a multivariate analysis. Such an analysis uses a variety of
techniques to determine important relationships amongst several variables.
While designing a research study, the researcher should give adequate thought to the use
of a particular analytical technique. In the recent years, many such analytical techniques
have proliferated due to the emergence of the computer. The researcher now has access
to an increasing assortment of techniques and it is desirable to know well in advance as to
what analytical techniques are going to be used, so that the data can be collected accordingly.
It is necessary that the researcher gives as much importance to the analysis and interpretation
of data as he has given to their collection. In the absence of proper analysis, data may be
rendered useless resulting in a waste of time and money.
a
K-P
reparing the research report
Once the data have been tabulated, interpreted and analysed, the Market researcheris
required to prepare his report embodying the findings of the research study and his
recommendations. A poor report on an otherwise good research will considerably undermine
its utility. Hence, it is necessary that the researcher gives sufficient thought and care toil*
preparation.
Although report writing needs some skills which can be developed with practice, tl
researcher should follow the main principles of writing a report. Some of these principle*
are objectivity, coherence, clarity in the presentation of ideas and use of charts and diagrams.
The essence of a good research report is that it effectively communicates its research
findings. As management is generally not interested in details of the research design and
statistical findings, the research report should not be loaded with such details, otherwise,
there is a strong likelihood of its remaining unattended on the manager's desk. In view of
this, the researcher has to exercise extra care to make the report a useful and a worthwhile
document for the management.
Sometimes, a detailed market research study throws up one or more areas where further
investigation is needed. Since research on those areas or aspects could not have been
fitted into the original project, a separate follow-up study has to be attempted.
^ Activity A;
Choose the correct alternative
1. The first step of Market Research Process is
a. research objectives
b. problem definition
c. collecting information
d. exploratory research
2. When one is seeking insights into the general nature of problem then we use
a. exploratory research
b. descriptive research
c. experimental research
d. panel design
3. The result of an explanatory research is

a. confusion
b. more information
c. research objectives
d. hypotheses
4. Sources of data for descriptive research does not include

a. interviews
205
Marketing Research
b. panels
c. surveys
d. experiments
5. Which of the following is an example of panel research design?

a. Census
b. TV channel ratings
c. Personal interviews
d. Survey
10.6 SUMMARY ____________________________________

The Market research process, as described above, involves various steps, though strict
adherence to each of these steps may not be necessary. A researcher may deviate from
the above sequence and steps depending on his specific needs. It should be remembered
that as research proceeds from the selection of the theme through the collection and analysis
of data to the preparation of a report, the focus of attention will move from one activity to
the other. This implies that the researcher does not always concentrate exclusively on one
particular phase of research until its completion.
Further, while it is beneficial to draw a detailed plan and sequence of various activities in
market research, it is hardly so if it requires such financial backing as the firm cannot
afford. There is no point in attempting something which cannot be completed on account
of financial constraints or limitations of time.
Another point worth emphasising is that howsoever elaborate a research design may he,
its successful implementation depends in no small measure on its management.
|
The market researcher should not only be familiar with the foregoing potential errors but;
should also make an earnest effort to have an effective control over themrHis effort should
be to ensure that his research study is free from these errors to the extent possible. This
means that he should be ever vigilant throughout the different stages of his research study
However, in practice, it is seldom possible to eliminate all potential errors. Efforts slMili be
made to minimise individual errors. If two or more individual errors occur, he should |
minimise total error by error trade-offs. Finally, he should measure or estimate the extent of
residual error that has still remained in the study.
Marketing Research
11.1 INTRODUCTION
All research activities need to address specified areas in which the research is to be
conducted. It also needs valuable information to be collected by means of a sponsor.
{ These are to be done on regular intervals. We shall discuss them in detail in this unit.
11.2 TYPES OF SPECIALISED TECHNIQUES

Certain specialised techniques of market research are used to address specific areas and
requirements of the sponsors. Instead of being one-time activities, these are continuous in
nature.
1. National Readership Survey (NRS): The aim of NRS is to provide information

acceptable to media owners, advertisers and advertising agencies as a third party
endorsement. These surveys are nationally accepted for identifying audience size.
reading habits and readership profiles, etc. specifically, the NRS investigates:
a) The readership pattern of all leading / major Indian publications i.e. dailies,
weeklies, fortnightlies and monthlies (magazines, newspapers).
b) The degree of duplication (cross readership) among various publications and

between various media. For example, a reader of Economic Times (Newspaper)
also reads Financial Express.
c) It covers cinema and TV viewership and radio listenership.
d) The socio-economic characteristics of readers / viewers / listeners of the various

mediums.
Education Profile Below SSC

ofreaders SSC Graduate
Post Graduates
"A">5 "B"
Readership by 1-5 lac
Town class "C"<llac
e) Readership by income levels

Unit 11 Specialised Techniques in Market Research
f) Readership per copy of publication
g) Readership by duplication
h) Rate of advertisement per column centimeter (Rs/cc) according to positioning

(advertisement appearing on front page, backpage, double spread, strip ads,
inserts, classifieds, supplements, etc.)
i) Readership by geographical area
With such detailed consumer and usage profile available with the media, planners can
plan efficient money spends to optimise results.
Retail Store Audit: It is a quantitative feedback technique from the market on

consumer buying patterns. The audit is primarily based on penetration, sales volume,
growth and distribution. A manufacturer is able to know what is the real off-take of
the product at a retail level. In India ORG-MARG does the retail store audit (RS A)
for fast moving consumer goods. The methodology uses a random multistratified
sampling.
a) Retailers are drawn from retail census classification from all over the country
covering major towns. The retailer's willingness to participate in the survey is
ascertained before making him a panel member.
b) Representatives of the research organisations call on these retail outlets on a
monthly basis to collect information on purchase, stock, sales, inventory, !
promotional schemes, etc.
c) Specially designed booklets are used to record various products, brand s, sizes,
etc.
d) The difference between the stock/inventory at the start and at the end of the
month is the sales for the said period.
Opening inventory + net purchase (receipt-transfer out-returned inventory)
-closing inventory + transfer in = Net sales.
e) Continuous monitoring helps to understand the market trends, depth of
penetration, promotional schemes, stock out situations, etc.
i) Sales quantity: Sales quantity can be shown by units, volume and
value
to consumers.
211
Marketing Research
Sales quantity = Opening stock+Purchases - Closing stock. ii)

Market Share: Individual product in ratio of total sales of the category.
Sales of product 100
Market Share % =
Total sales
in) Growth: Refers to the percentage change in current sales over the
same period in the previous year.
iv) Stock turnover Ratio (STR): Indicates the number of days the stock is
going to last given the current offtake levels.
Stock
STR= --------x (30 days)
Sales
v) Per dealer off take (PDO)

Stock by volume
PDO =
Total no. of dealers vi)
Out of stock Dealers (OSD)
Out of stock dealers 100
OSD%= --------------------------
Total Dealers
f) Retail store audit is made to be user friendly. It divides India into four zone;
(north, south, east, west) and 16 states. The sample spread is 8800 retailers
across 367 towns (all India) out of which 345 5 retailers are urban area retailers
and the rest are in rural areas.
g) Monthly report on urban/rural India and all four zones is given to all its subscriber!
h) New product introductions, promotional schemes and pack variants, dealet

I incentives are easily traceable through this data.
There are, however, certain limitations in this method. They are as follows:
It requires a large trained manpower for data collection and electronic
ffi< processing techniques.
I
: • Not all type of retail operations are audited. Certain retail outlets do not
co-1 operate in data collection.
*{•
• Brand loyalty cannot be known through this data.
1 Consumer Panels: This method is quite popular abroad but has not been a
much-favoured method in India, though IMRB has household panels to provide
purchase specific data in select metres. Consumer panels help obtain qualitative
feedback, especially the brand loyalty of consumers. Consumer panels are
selected through stratified random sampling so that the sample is the representative
of the class. The size of panel depends on the nature of the study. Members are
required to maintain a diary recording all purchases of regular products or research
products, size, price, the units bought, the place of purchase, sometimes even the
reason for purchase. The reporting could be weekly, monthly or as desired by the
sponsor of the study. The data is then gathered and analysed to obtain useful
information on retail sales and its trend, purchases by existing and new customers,
brand loyalty, market share, effect of premiums and discounts on sales, kinds of
stores in which the purchases are made,
! effect of price change, socio-economic classification of purchases and the members
buying multiple brands for a single product.
An adaptation of this technique is called 'Brand Barometer' in which panel members
are retailers instead of consumers to give information on brand loyalty.
4, Stimulated Test Marketing (STM): In STM, market-like conditions are stimulated
and the selected sample base is studied for trial and repeat purchases in an attempt to
estimate the market share. This technique is known for its versatility. Selection of the
sample unit is random. A sample area which is selected is most representative of the
entire market. Time and cost considerations also guide the selection process. Free
samples with or without the brand name are distributed and consumers are motivated
to use them. Then their opinion after usage is recorded on product attributes, packaging
quality, intentions of purchase /repurchase etc., through a structured questionnaire.
Buying is a complex process involving social and psychological factors. Through test
marketing, the risk of product failure is attempted to be minimised. The test marketing
is done at two stages.
a) Pre-Productfinalisation
b) Post-Product finalisation before a nation-wide launch
213
Marketing Research
Simulated test marketing is done for the following purposes:
a) Check product formulations so as to choose between alternatives
b) Brand names and Brand recall
c) Pack size and Packaging appropriateness 2.
d) Product concept testing
e) Product test with advertising
f) Product test with prototypes
g) Trial, repeat and frequency of purchase
h) Pilot study before national launch
11.3 RESEARCH IN ADVERTISING DECISIONS

Promotion of products and services through advertising is becoming a very costly exercise
An advertiser is always wanting to gain mileage through every rupee spent so as to comer
it into sales.
Some of the aspects that are studied through advertising research are:
1. Target audience characteristics and size
2. Copy testing and advertisement lay out
3. Media research anditiselection
4. Corporate image study^^
5. Consumer attitude measurement^
6.
7. Advertising Agency
evaluation The two methods of
research are Recognitio
• Opinion Research n Recall
• Memory Tests
Specialised Techniques in Market Research
1. Opinion Research: This methodology provides for mere feedback expressed as

opinion on advertisement copy, design and layout after the respondent is exposed to
advertisement. Opinion surveys are qualitative measures and cannot be quantified.
Aspects like media recall, how long it appeared cannot be measured.
2. Memory Tests: Memory tests are an improvement over opinion research. There
can be variations in the test. They are administered after the release of an advertisement
campaign.
a. Recognition test : Showcards are developed containing the clippings of

advertisements. The brand name may be purposely deleted from the
advertisement. Respondents are then asked to identify the brand and the media
in which it appeared.
b. Recall test: In this method respondent lists unaided all the advertisements seen
in a particular product category. For example, all TV commercials seen for
soaps. High recall is a sign of superior attention grabbing power of the
advertisement.
The respondents may also be asked about seeing a particular advertisement.

e.g.
Can you recall the Liril ad in which model x is seen?
3. Media Research: An advertiser has the option to choose from various media like
the TV, press, radio, cinema, etc. for the release of an advertisement. With a limited
outlay, it is imperative to derive marginal utility from every rupee spent. Each medium
is tested on its versatility v/s cost before taking a decision. The key research aspects
klude
a. The efficient media to reach the target population. Which medium offers the
maximum reach in the target audience of the specific product?
b. Cost effectiveness with respect to the audience media characteristic.
c. Demographic characteristics and the market potential of the target group.
d. Feedback on the noticeability of various advertisements in different media.
4. Point Of Purchase (POP) Research: POP items are widely used by all marketers.
They are important communication media that help a consumer to recall advertising
or stimulate consideration to buy the product. Various POP items that can be seen in
Marketing Research
day-to-day life include posters, stickers, backing sheets, danglers, buntings,

neon-signs, glow-signs, tinplates, balloons, hangers, stands, banners, floor prints,
enlarged packs, leaflets, shelf talkers, dispensers, audio voice overs, trolley plates,
etc. The list is never ending as marketers are developing newer methods to catch
attention amongst the clutter of brands and space.
POP is researched through Forced Exposure Test (FET). The respondents are shown
POP under a controlled stimulated environment and their response is recorded for
I
the following:
a) Attractive and attention catching
b) Standout factor amongst POP clutter

c) Message and content comprehension
d) Brand linkage
e) Uniqueness
f) Overall likeability/dislike
g) Installation difficulties
Effective and innovative use of POP items can at times work better than adverte 1MIIJ
Opinion Polls: Opinion polls are often used to cover a variety of topics m
ranging
political, social, legal, economic and educational issues. The sponsor of the opinion
poll needs to identify the topic for poll, area coverage, target audience, time ami
budget constraints and the kind of data analysis required. The simplest of opiraaf
polls are the Yes/No/Can't say answer types. TV programmes, SMS (cell phooaj
services and internet sites are using a lot of opinion polls on current issues on adii
basis. i
11.4 MOTIVATION RESEARCH
If the marketer got to know the reason why people behave the way they do. t
problems would have been solved. The decision making process is a complex one
motives guide a person towards action.
M
A motive may be defined as an inner state that energises, activates, directs the belli alJ pa:
of an individual towards certain goal or activity. Vai
«*'•
There are various factors that affect the consumer decision process. They are as follows:
a) Economic factors
b) Intrinsic factors (pertaining to personality individual make up)
c) External factors
In recent years, marketing researchers have borrowed concepts from psychology to

understand and measure emotional factors involved in a consumer buying decision.
Motivation research is based on two basic assumptions.

i Every individual has drives which when stimulated, generate a response.
• Every individual is influenced to a certain extent by his environment.
Motivation research is difficult to handle as it deals with qualitative variables present in

human beings who are dynamic by nature. Consumers are generally not aware of the true
reasons for specific brand/product selection. Sometimes they are unwilling to tell the truth
if it pertains to sensitive issues, moral issues, or if they have inadequate knowledge. For
example, views on drinking, sex, etc.
11.5 TECHNIQUES OF MOTIVATIONAL RESEARCH __________________
Techniques of Motivational Research
Attitude research Depth research Protective research
liken Semantic Indepth

1
Focus Focus • Word asso
Error differential interview group group
choice scale interview• Sentence completion
Story completion
Rosenweig test TAT
Doll play
Motivation research helps identify why the same stimulus fails to bring the same response
ithe time even with the same person. Various intervening variables like personality, attitude,
KM experience, information and evaluation criteria differ from person to person. These
variables (motives) that cause a response exist at three different levels of awareness.
21
7
Marketing Research
Attitude Research : Attitudes are predispositions towards products and services.

Verbalisations of these attitudes are opinion. The measurement of attitudes is done through
scaling techniques. Commonly used attitudinal scales are:
1. The Likert scale: As dealt in the chapter on scaling.

•s^
2. The error choice scale: In this scale, the respondent is given a question and is then
asked to select an answer from a series of probable answers for which little factual i
information exists with a respondent. For example,
Q. The average net profit margin for Procter and Gamble is: 2%, 3%, 4%, 8'.
10%, 15%, 20%, 24%, 28%, 40%, 45%. The answer will reveal respondents
attitude towards them. Do they think that the business unit is profit oriented ot I
consumer oriented?
3. Semantic Differential scale: As dealt in the chapter on scaling.
Depth Research: Depth research techniques are used when respondents are unwilling
or unable to respond meaningfully to direct questions.
The idea is to get consumers to talk all about their feelings concerning a product the
pleasures, joys, agonies, enthusiasm, nightmares, deceptions, apprehensions and so on
that a product brings to their mind. Such researches are conducted by psychologists rate ]
than a marketer.
The process involves:
1. Establishing a rapport with respondents
2. Indirect questioning
3. Stimulating response through probing and empathising, for example, ah,

oh.ys
what more? etc.
4. Noting the reactions of the respondent.
Through this method, unexpected motives may be discovered. However, this is

not with a large sample size due to time, cost and interpretation constraints.
Focus Group Interview: In this method, a small number of consumers or potential

(5-10) are brought together to informally discuss a marketing situation. The group steers
the discussion to reach a conclusion. However, individual freedom to re,
Marketing Research
An inkblot similar to the Rorschach blots
Thematic Apperception Test (TAT)
Commonly referred to as TAT. This test consists of a series of pictures about whidifc
subject is asked to tell stories. These pictures may deal with day to day ordinary euw
may represent unusual situations. The respondent is encouraged to tell -
a) what he thinks led to the scene/situation
b) what is happening and the feelings of the characters in the pictures
c) what the outcome will be. The responses are recorded verbatim and the inve
draws certain inferences about their personality structure, tensions, adapt Uîi
expressive behaviour. It is assumed that the respondent's externalisation is the

of the internalisation of his self.
1
220
Example:
Q. Who do you think i s more healthy?
Ans. To be given by the respondent by looking at the picture.
Doll Play Test
This test is especially suited for children. Dolls representing adults, children, ethnic races,
both the sexes, are given for free play to children. The children may be asked to show how
thesedolls will act in specified situation. The manner in which they organize the dolls and
play with them represent their attitudes, prejudices towards the object/situation represented
bythedoll.
The Rosenweig Test ,
Ituses a cartoon format in which one character is represented as saying something. There
isablank space for another character and the subject is asked what this second character
raldprobably say. Marketing situations may be studied through this techniques.
Marketing Research
The Story Competition Technique
In these
tests, the
subject is
given just
enough of
a story or
argument
to focus his
attentsr
on a given
issue but
not enough
to indicate
how it will
turn out.
The
respondent
is ;ii;
provide
the end /
conclusion to the story. t
The way the respondent supplies the ending statement is assumed to portray hb
fc seated biases. For example, Mr. and Mrs. Desai purchased a food processor. The\
^ the sealed box to find some spare parts missing. They went to the retailer...
Word Association
las been dealt under the chapter, Questionnaire, subheaded under open ended questions.
Sentence Completion
Asentence is read/given to the respondent who is asked to complete the sentence with the
first thoughts that occur to him.
Forexample, I Love Diwali because
Responses could be
i we get to go home for a holiday

M+ • - .
- a student
i
• we eat sweets and burn crackers
-a child
t it marks the beginning of a new year
'r -a housewife
• we get bonuses to buy new clothes and sweets
- a worker
i there is fun and festivity all around
- a youth
JfWI"
Each statement is marked by responses to things that are most valuable to each one of
lea
11.6 LIMITATIONS OF MOTIVATIONAL RESEARCH _________________

1, Motivation research depends on various techniques borrowed from other fields. The
transfer of techniques from one field to another poses problems.
1 Thecostof interviewing is generally large.
Marketing Research
3. Interpretation of data by different people doing the same test poses the problem c
subjectivity.
4. The validation of results is impossible.
^ Ac tiv it y A:
a) Two children were quarrelling at the breakfast table. The elder boy spilt milk on his]
uniform. It was time for the school bus to come. What did the mother do?
b) What strikes your mind when you hear the following words?
Black ___________
Brave____________
Cunning____________
Honest _______
11.7 RESEARCH FOR PHARMACEUTICAL PRODUCTS
Over the decades the pharmaceutical industry has seen a radical change. Phai
manufacturers are now trying to reach the masses for brand awareness and the comp
reputation so that the prescriptions generated are not substituted at the chemist level.
the companies are making rapid strides in manufacturing and exporting the erstwhile patent
drug. With this need for money through public sourcing, the company image becomes!1
necessity. . •
The covert image of this industry has changed and pharmaceutical marketers are iBif
T
market research for various reasons.
Ti
Specialised Techniques in Market Research
PHARMA COMPANY
1 I jrmulations
i Market
' ing needs
Prescription Generic
t * 1 |
i
ical PerformanceI Packaging uct Product Prescription r i r
Proc als tracking ept research audit Ret aui ail
cone lit
Pricing
OTC Quantitative Company
Qualitative (Over the counter) image and
Product awareness
Brand concept
Packaging image
SOURCES OF INFORMATION
1 1
Medical Pharmacists Chemists Patients Public
Practitioners
11.8 THE SOURCE OF INFORMATION FOR PHARMA PRODUCTS

Itistobe noted that people like doctors, pharmacists are well trained; they are educated
people. One should keep in mind the following:
» who understands technical jargon
• are busy people
• every minute wasted is a money earning opportunity lost
• do not like to deal with unsophisticated and unqualified professionals
• research on his patients is a learning experience
• also cannot risk a patient's life due to medical ethics
lteefore,the person conducting the research should be well qualified so as not to waste
Moctor'stime, as well as to settle his product queries regarding various scientific issues.
Ike various issues which a doctor seeks regarding a product are as follows:
225
Marketing Research
1. Dose and dosage schedule: Amount of the drug required and at what intervals.
2. Indications for prescription
3. Overall efficacy, the time needed for drug concentration in the body
4. Drug clearance time from the body
5. Drug interactions (if any) for example, not to be given concurrently with calcium
carbonate
i
6. Side effects for example, dizziness, metallic taste, etc.

I
7. Contra indications for example, not to be given to children under 5 years, pregnm
and lactating mothers, the first trimester of pregnancy, people allergic to sulphides
etc.
8. Precautions (if any) while administering the drug
9. Overdosage - Remedies and accidental damage

10. Product availability forms: Tablets, capsules, cream, injection, etc.
For research . . . . . < t t...

Target respondent is selected: Which category of doctors? For example, drug wfllk
used by a gynaecologist, general practitioner, pediatrician, etc. Also the ability to influenct|
practitioners is seen.
The nature of practice and the kinds of cases dealt by the respondent doctor is c so
as to deal with all the indications of the product.
Sample size: To be representative of the population
Survey centre: Places, hospitals, etc. are specified
Sampling procedure: Area sampling Research
process
Thi
• The target doctors are visited for prescription information needs. pha
• They are then exposed to the product concept and the product details.
• Participation and cooperation in the trial is solicited.
• Each participant doctor is provided with adequate samples to cover participating

patients for the complete prescription duration.
• Regular visit or monitoring for the proper conduction of the trial.
« Feedback in structured/unstructured questionnaires for product results
MRB has established Pharmatrack to handle a specific market's requirement. The retail
store audits and prescription audit are carried out by ORG on a continuous basis just like
FMCG products at all chemist and stocking points to compare the competing product
offtake. Pharmatrack reports are confidential and not to be shared whereas ORG audits
are given to all its subscribers. Some areas of research are
1. Prescription patterns and factors influencing prescriptions
2. Proportion of doctors detailed (told about the product) prescribing the said brand,
the alternates prescribed
3. Reaction to the new products
4. The drug and the company image

Ik
5. Reasons for brand switching or not prescribing
6. Prevalent cases (ailments)
Over the counter drugs (OTC)
These are medicines which the consumers buy on their own without prescription because
of the safety of usage in the drug category. These products can be advertised into various
mediums like TV, press, radio, etc. For example, Anacin, Moov, lodex, Sandoz Calcium,
Krack, etc. The efficacy of the product is checked by the manufacturer himself. For all
other aspects like packaging, pricing, image, brand recall, sales audit are treated like all
FMCG products as they are available at all types of retail outlets.
This unit is an attempt to give a bird's eye view of research utilisation in the field of
Pharmaceuticals.
227
Marketing Research
11.9 RESEARCH FOR INDUSTRIAL PRODUCTS

Traditionally, the industrial segment has had limited customers and a selected range of
products offered by companies. With globalisation, various technical and chemical
innovations, the product portfolio of companies is going through a sea change. To remain
competitive in the field of limited scope, a lot of research projects are now getting initiated,
This was earlier restricted to areas like demand forecasting only. Avenues for research
include:
1. Company Image studies: These measure corporate image at a particular point of
time or they could be done on a continuous basis.
I
2. Brand image : This may relate to the company or to the product as brand. The '
attributes and intangible benefits associated with such products are
3. Pricing research : This is crucial for new product introductions or where the
competition is stiff from the unorganised sector as well.
*
4. Preference studies: These are attitudinal studies towards a particular product or
service. «
5. Buying influencers : These studies determine the various factors influencing

organisational purchase behaviour. ^
6. Product specifications studies: The user's needs and the competitive machine?
are studied in depth for new product introductions.
7. Technology innovations: The new technologies have to be studied and adaptabft
to be determined for industrial usage. For example, the equipment using solars
wind energy. k
8. Substitutability of new materials: Newer raw materials are studied for priciit
and functional advantages. For example, new polymers are replacing steel and plasw
11.10 SOURCES OF INFORMATION FOR INDUSTRIAL PRODUCTS

The sources can be collected with the help of:
Primary sources Secondary sources
Through the
tie above resources, both, qualitative and quantitative aspects about a product
can be studied. For demand estimation studies, factors like the life of a product, the
replacement market and international scenario (imports etc.) should also be considered.
Details about the methodology has been discussed in previous units.
11.11 SUMMARY ______________________________________________

We have seen that all the techniques need data and information. These can be collected on
aregular basis with the help of clear sponsorships. We have also seen that advertising
decisions play an important role on the basis of the different polls on which the ads are to
be made, which gives the different methods viz, the opinion poll or the copy test or corporate
image study. They are to know how people react to these. Retail store audit gives a clear
picture on the sales quantity per quarter and the market share of the product.
11.12 KEYWORDS ___________ ' _________________________________

Intrinsic: inherent, essential
Jargon: word or expressions used by a particular group or professionals
11.13 SELF-ASSESSMENT QUESTIONS _________________ _ _ _ _ _ _ _ _

Ql. What is a retail stores audit? Elaborate your answer.
Q2. Explain the national readership survey. Elaborate your answer by considering your
own example of any survey you have so far seen.
Q3. What are the major aspects that are studied through advertising research?
Q4. Write a short note on POP research.
Q5. Give some situations where you would like to use stimulated test marketing.
229
Marketing Research
12.1 INTRODUCTION
The marketing management performs the vital functions of policy formulations, planning
marketing operations and devising methods to control marketing efforts including sales so
as to achieve the organisation goals. For the efficient discharge of these functions, the
management needs complete information about the market and sales - present and potential.
Thus, the management has to heavily depend on research in these areas. Market and sales
analysis which covers the identification and measurement of all these variables individually
and collectively have an effort on sales.
To be more specific, market and sales analysis research includes:
• Measurement of the market - present and potential,
• Demand or sales forecasting,
• Sales analysis territory-wise, product-wise, salesman-wise, etc. and also
• Evaluation of the performance of the sales force.
The importance
Market and sales analysis is of paramount importance because it helps the management in:
a) Developing a company's sales plan (short and long term).
b) Estimating the future sales.
c) Establishing sales goals for each unit and for each individual in the sales organisati
d) Setting standards of sales performance.
e) Allocating a salesman's time, efforts, territory, etc.
f) Estimating the total sales force needed in the future.
g) Taking decisions about the remuneration (commission) of Ifehe salesman, training E

compensation and,
h) Preparing a sales expense budget.

Unit 12 Market and Sales Analysis Research
Most of the marketing research people are engaged in market and sales analysis research.
Market analysis refers to those research activities which attempts to determine the sales
potential. It includes sales potential for the total industry and sales potential for an individual
firm. Sales forecasting would include:
1) Estimating sales for a particular product
2) Company
3) Branch
4) Unitor
5) Department for a particular period of time
12.2 TYPES OF SALES ANALYSIS __________________________________

Sales analysis will include studying of sales - territory-wise, type and size of customer
order size, product-wise and combination of these.
Market Analysis: Marketing management is interested to know the total market for a
product in particular. In other words, if the management is aware of the total potential of
sales by all firms in the industry, it will be able to plan well about its own share. The total
market for a product or service can then be determined geographically, territory-wise or
area-wise. While estimating the total market for a product, the research people must take
care of the substitution of product 'X' to 'Y' and vice-versa by the consumers. The total
market potential can be estimated with a different set of assumptions. For example, the
market potential for say, toothpaste would be of one order if we assume that people brush
once in a day and the market potential would be different if we assume that people brush
their teeth after every meal. The total market potential is of great use in the following ways:
*
Helps in defining sales territories
As a sales manager wants to develop sales territories that are nearly equal in work-load so
that the sales force can be effectively used.
A sales representative's work load depends on
a) Number of accounts or clients - distributors or dealers
b) Size of the accounts
233
Marketing Research
c) The average order size ;

d) Location
e) Travel time between two clients
f) Non-selling activities required of a salesman (filling up daily contact form, gathering
data, commenting on the competitor's activities, details of accounts/recoveries with
the dealers, etc.)
If a company is able to know the market potential, area-wise, it can successfully plan the
number of people to be deployed to each territory. It would also help the management in
deciding how much investment is to be made in the development of that territory. Money
to be spent on advertising and non-advertising promotional methods can be decided on
the basis of the potential. If Bombay has six times the potential than say, Poona, resource
input should be accordingly planned. A company operating nationally may advertise on a
national media, but may go for local advertising depending on potential. In this way the
key markets are carefully served.
In addition to potential, data on the competitive structure and strategy should also be
collected. The firm's own strength must also be considered.
Setting sales Quotas
Once market potentials are estimated and sales territories established; the company can
go for setting the sales quotas. Sales quotas must take into consideration the past
performance in case of an old company. Traditionally, sales quotas were set up by rule of
thumb and if a salesman was not able to do the desired task, he was fired. Take another
situation where 'A' and 'B' contribute equal sales from the territories 'X' and 'Y' respectively.
The sales manager may view them as equally efficient. But the fact may be that if territory
'X' has a high a potential than that of 'Y', they should not be considered equal.
12.3 AREAS OF SALES ANALYSIS

Sales analysis is another important area needing research.
Sales Analysis is done in four areas :
1) Territory
2) Product
3) Customer
4) Order Size
|. The invoice is usually the basic document. An invoice contains the following:
-4
a) Party's name
,-, b) Product sold
c) Quantity
d) Price
e) Total Sales or Order
Based on these invoices territory-wise, the sales break up can be studied. Similarly, the
sales analysis by product line or if an individual product can be carried out. Product decisions
can be taken only when we have sales analysis product-wise.
Customer-wise break up of sales may reveal that a small percentage of clients contributes
to a large percentage of sales. Such customers must be taken care of.
les Analysis by order-size may reveal the points of good sales volume.
Combined with all the 3: i.e.

a) Territory
b) Product and
c) Customer-wise
Sales Analysis may reveal the weak and strong points in sales programme.
12.4 METHODS FOR MEASURING MARKET POTENTIAL

There are two methods to measure market potential. They are as follows:
1) One is through direct data i.e. data on the actual product for which we want to
estimate the potential.
2) The other method is through the use of corollary data i.e. data related to the product
but not about the product.
Marketing Research
1. Direct data method
The total sales of a commodity can be estimated for the entire industry. Its
territory-wise break up is obtained. The territory's characteristics are taken into
account along with the past sales, the percentage wise distribution for the industry sales
with territory-wise provide a base for estimating the market potential for our brand. The
comparison of actual sales and potential sales will tell us about our weakness or
strength in that territory. There is a disagreement whether a company should
concentrate on a weak market or on a strong market. A company, however, must
carefully decide the action to be taken in each territory. Sometimes, the total industry
sales can be easily estimated according to each area if the product is under license or is
regulated through institutions like the State Trading Corporation or the Minerals and
Metals Corporation (MMTC) etc. The federation of Indian Chamber of
Commerce and Industry (FICCI) and many other associations concerned with each
industry compile such data, like a product by the Indian the Jute Manufacturing
Association (IJMA).
It is evident from the above that past sales form the basis of measuring potential. But
past sales are always a function of advertising, price and the promotional effort put in.
These variables may change the sales potentials.
2. Corollary data method

:
'' !
Measuring the market potential is based on the idea that data related to one product :
would automatically tell the sales potential of another. For example, automobile parts :.
(spares) replacement may depend on the number of automobiles in a particular region, r
Further, the need of tyres may directly depend on the market potential of that vehicle. ,
Population on households may also be used as a base for measuring that market f
potential if the product is directly related to all households irrespective of their status •
economical or social for example, the consumption of salt. If one area has twice the I
population of another area, then it has double the potential of the other area. However.
one should never ignore the weaknesses of the method adopted. For example, its J
population alone may not indicate the total purchasing power or availability of!
discretionary income. In fact, more than one series of indexes may be used to estimate'
the market potential.
Use of surveys to measure market potentials
From a sample of respondents, one can determine the amount of purchases. Using tk F
sample data, the average purchase per person can be estimated. Multiply this average h the
number of people in that area.
SALES FORECASTING
Most of the market planning is based on its sales forecast. In the present environment,
Is are produced in anticipation of the demand. Therefore,
a) purchase of inventory
b) production
c) manpower requirements
d) advertising expenditure and
e) cash flow needs
are some of the areas where decisions are based on a sales forecast. Many companies
make several plans based on alternative forecasts.
Sales forecast is a complex subject which uses a variety of concepts and techniques. Sales
forecasting methods are of two types namely.
1) Subjective methods
2) Objective methods
Forecasting
•
forecasting is an attempt to see the future by examining the past." -Alderson and
Green
Itinvolves the determination of the expected values of some variables in the future based I
on past and present behaviour, the management's intentions and the constraints imposed on
the enterprise. It is an objective and systematic activity different from mere guesswork. Itis
quite complicated yet inevitable because it is an integral part of planning. It is not only
aplanning device but a controlling tool also. Even if it is imperfect, it acts as a monitoring
device during the implementation phase.
m . .. .
12.6 METHODS OF SALES FORECASTING ________________________
A. The diagnostic methods of sound managerial judgement or the intuitive method. Under
this method, the research analyst is not concerned with the cause and effect relationship.
All the methods under this group go by the intuition of the expert.
237
Marketing Research
Various methods under this category are as follows:
• Munshi Forecast - (short term)
• Lost Horse Method
• Delphi Method (in the long run)
1) Munshi Forecast
Scientific forecasts may not be so accurate as the Munshi forecast since he

collects all information in his mind and based on his vast experience, he makes a
forecast.
2) Judgement method or the Lost Horse method
Some analysts may depend on value judgements by some expert. This is also
called the Lost Horse Method. The basic idea is to set up the position of a
person who has to make a decision regarding the direction in which the lost
horse might have gone. Information on national income, consumption, investment
import export may be collected and plotted on the graph which will provide
direction. By sheer inspection, one may decide the future direction.
3) Delphi Method
Experts or knowledgeable people are invited and they give their forecast without
giving reasons. A questionnaire may be used for this purpose.
B. Barometric Methods
1) Historical analogy
2) Product life cycle concept
3) Survey methods
a) leading indicators
b) coincident indicators
c) lag indicators
238
1) Historical analogy
»» For macro level forecasts, we try to find out a period in history where conditions
similar to those prevailing in our country were present in the past. Say conditions
in 1890 in USSR, USA, 1910 in UK, 1920 in USSR may be same as in the
1970's in India. This may give an idea of long-run forecast. Though it is not a
very sound method, it is used for national forecasts.
2) Product Life Cycle Concept

*• Product Life Cycle Concept is also used to make a forecast about the
product sales in the future.
3) Survey Method
Through a survey, data may be gathered which may act as an indicator.
Economic indicators are statistics that tend to increase or decrease in value

ahead of or at the same time or after some time. The choice of indicators is
crucial. Indicators may be of three types namely
a) Leading indicators or advance indicators move in the same direction as the

series being forecasted but they do so ahead of times. For example, the
t demand for transformers can be forecasted based on the indication of house
I activity in a particular region.
b) Coincident Indicators move in the same direction at the same time like
employment and industrial production.
c) Lagging indicators move in the same direction but behind the series for
example, unit labour cost.
The National Bureau of Economic Research publishes data for coincidental

lagging indicators and leading indicators based on 350 time series.
Trend Analysis through Extrapolation

This method is concerned with the extrapolation of the series based on past behaviour.
•if y=fxs x c x i, where trend = long term movement, seasonal, cyclical and irregular.
239
Marketing Research
Regression Analysis
Regression analysis can be used in sales forecasting to measure the relationship between a
company's sales and other economic variables. For example, an automobile manufacturer
may find that the sales of his car is dependent of the income level. In other words, as the
income level goes up, the sales goes up.
Regression analysis is one of the most frequently used techniques in social science research.
It is often used when the researcher is interested in estimating or predicting the value of
one variable, given the value of another.
For example, it is believed that advertising expenditure and sales are related in such a way
that the former increases sales. In such a case, one may like to know the likely sales
against a given advertising expenditure or vice versa. A regression analysis is used for this
purpose.
The relationship between the two variables can be either linear or non-linear. When the
relationship is linear, a change in the independent variable is followed by a constant absolute
change in the dependent variable.
The linear relationship between the two variables x and y is of the form Y =
a + bX
where Y is the dependent variable and X is an independent variable. It is conventional to

use the terms 'independent variable' for X and 'dependent variable' for Y. This use does
not imply that X depends on Y or Y causally depends on X. The regression model does
not necessarily imply any cause and effect relationship between the two variables. However.
the model implies that a change in the value of X will bring about a change in Y such that
the pair of observations, if plotted on a graph will form a straight line. The regression
equation indicating the specific relationship between the two variables can be found out
with the help of normal equations. This is illustrated by the following example. A trading
company has the following territory- wise data with respect to sales and advertising
Sales Territory 1 2 3 4 5 6 7 8
Advertising ('000 Rs) 40 30 20 50 60 40 20 60
Sales (in units) 100 80 60 120 150 90 70 130
240
The table 12.1 below will give us the calculations
Table 12.1
Sr. Sales Sales (in Advertising

No. Territory units) ('000 Rs)
Y X X2 XY
1 100 40 1600 4000
2 80 30 900 2400
3 60 20 : 400 1200
4 120 50 2500 6000
5 150 60 3600 9000
6 90 40 1600 3600
7 70 20 400 1400
8 130 60 3600 7800
800 320 14600 35400
2
E X = 320 E X = 14600
EY=800 E XY = 35000
The linear regression is of the form
Y=a+bX
where Y is sales and X is advertising, the parameter a is the intercept and the parameter b
is the coefficient of variable X. The two normal equations for fitting the regression line
are:
Substituting the values calculated above in these normal equations,
lij 800 = 8a + 320b
!n) 35400 = 320a+14600b
241
Marketing Research Unit 12
Multiplying (i) by 40 and then subtracting (ii) from the resultant figures,
Cm) 32000 = 320a + 12800b (iv)

35400 = 320a + 14600b
- 3400 = - 1800b 4
therefore b = 3400/1800 = 1.8889 approx

'' ' 4
Substituting the value of (b) = 1.8889 in (i) above
800 = 8a +(1.8889x320)
or 8a = 800 - 604.448
Ia=
24.444 approx
Hence, the regression equation is
Yc = 24.444 + 1.889 X (Rounding the values to 3 decimal places) '_•
In the above calculations, we have used the normal equations to estimate the parameters
a and b. Alternatively, the values of a and b can be obtained directly by applying the
formulae.
Parameter bean be computed by the following formula: {•
b= n
Applying this formula to the data given in Table 12.1 we get
8(35400)-(320)(800) *
Hence, the regression equation is
Yc = 24.444 + 1.889 X (Rounding the values to 3 decimal places)
In the above calculations, we have used the normal equations to estimate the parameters
a and b. Alternatively, the values of a and b can be obtained directly by applying the
formulae.
Parameter b can be computed by the following formula: f;
b= n(£XY)-(£X)(£Y) 3
Marketing Research
independent variable. A high value of parameter b shows its importance in the regression
equation.
The parameter a is the Y intercept of the regression line. The magnitude of this intercept
shows the value of the regression function at X = 0. When a is 0, the regression equation
of Y on X will pass through the point of origin where both X and Y are 0. Alarge intercept
suggests that the dependent variable Y is determined by something other than the
independent variable X. In such a case, the researcher should look for one or more other
variables, the inclusion of which would improve the regression model.
A market researcher who uses the technique of regression analysis would be interested in
knowing how good it is. In other words, how much variation observed in the dependent
variable is explained by the regression equation. For this purpose, he is required to calculate
another measure known as the coefficient of determination which, in fact, is the square of
correlation.
The coefficient of determination (r2) is calculated as follows:
(Total variance in the dependent variable) - (Variance "unexplained" by the regression

equation) »
Total variance in the dependent variable
—\2
Symbolically, r2 =
where Yc, a new term, is calculated by applying the different values to the independei
variable X, as given in the problem.
Taking our example where the following regression was arrived at
Yc = 24.444 + 1.889 X
The vali
about
Table 12.2 gives the calculations for determining r2
9-5.6 per
(
regressi
Inoure>
relation*
Table 12.2 : Calculations for Determining r 2

X Y Y.-Y (Y.-Y)* Y Y.-Y (Y.-Y )2
c 1 C v , c /
40 100 0 0 100 0 0
30 80 -20 400 81 -1 1
20 60 -40 1600 62 -2 4
50 120 20 400 119 1 1
60 150 50 2500 138 12 144
40 90 - 10 100 100 -10 100
20 70 -30 900 62 8 64
60 130 30 900 138 -8 64
800 6800 378
= — = 100
—\2
= 6800 - 378
6800
= 0.944
The value of r2=0.944 shows that the regression equation Y=24.444 +1.889 X explains
about 94.4 per cent of the total variation observed in the dependent variable. Thus, only
5.6 per cent of the total variation is the dependent variable, Y, remains unexplained by the
regression equation.
Inourexample, r2 is extremely high, though it may be sometimes small. Acertain degree of

relationship between two variables may occur on account of random sampling variation. It
245
Marketing Research
is, therefore, desirable to test the hypothesis that r2 = 0. Let us illustrate this with respect to
our example.
The null hypothesis, 1^:^ = 0
The alternative hypothesis, Ht: r2 ^ to 0
A t-test is used to test the null hypothesis. The formula used is:
_ Wn - 2
t— y with n - 2 degrees of freedom
Applying this formula to our example
0.97 >/8- 2
Vl -0. 94
_ 0-97 x
2.45 0.24
= 2.38/0.24 =
9.92(approx)
The critical value of t at « = 0.01 for 6 degrees of freedom = 3.707. Since calculated^
exceeds critical t, the null hypothesis is rejected.
In other words r2 ^ 0
12.7 SUMMARY______________________________________________
When we are interested in measuring the degree of relationship between two variables
we use another concept, i.e. coefficient of correlation. The two terms correlation n
regression, sometimes used interchangeably, are distinct. Correlation is a statistical techiijiit
used for measuring the relationship or inter-dependence of two or more variables, nonet:
which is restricted by the researcher. Correlation does not necessarily indicate a caua
relationship between two and more variables. ||
Regression analysis, on the other hand, refers to the technique for deriving an equationttoP
relates the dependent variable to one or more independent variables. It is used to prei one
variable on the basis of another variable. It is also helpful in bringing out the cans
246
relationship between two or more variables.
Whenever a correlation analysis is to be attempted, it is advisable first to draw a graph of

the two series in question. Such a graph is known as a scatter diagram wherein the values
of the two variables are plotted in such a manner that a point on the graph indicates a pair
of values. The usual practice is to plot the independent variable on the horizontal axis and
the dependent variable on the vertical axis. A scatter diagram reveals whether the movements
in one series are associated with those in the other series. In case of a perfect correlation,
the points will fall on a straight line in a diagonal form. If this straight line is rising on the
right, the correlation is positive and if it is falling, the correlation is negative.
12.8 KEYWORDS ____________ - ______

.
Vital'.essential
Corollary: proposition that follows from one already proved
12.9 SELF-ASSESSMENT QUESTIONS______________________________

Ql. What is regression analysis?
Q2. What are the different types of sales analysis?
Q3. Elaborate the different types of sales forecasting methods and illustrate your answers
with examples.
247
Marketing Research
13.1 INTRODUCTION
A new product should be developed before it is launched. It is very necessary to test the
product in the market for various reasons. Once the product is accepted by the target
customer, then we say the test marketing has been successful. In this unit, we are going to
discuss in detail about the new product development, test marketing and advertising
research and why research is important for advertisement.
13.2 NEW PRODUCT DEVELOPMENT

A company may find itself in a situation where it may be advisable to develop a new
product. When sales of its current range of products have been declining over the past few
years or when it receives complaints about its products from customers, distributors.
retailers, it may have to find the possible reasons for the same. This may lead to the
improvement of the existing product or the development of an altogether new product. As
we know, the concept of product life cycle suggests that a product passes through different
stages.
When a company finds that some of its products have entered the declining stage, itrna;
have to take concrete measures to replace them. This can be achieved in two ways
acquisition and innovation. We are concerned here with the latter which can be broadly of
two types: internal innovation and contract innovation. 1)
New product development can be carried out in one of the following ways:
New product features can be developed by adapting, modifying, magnifying, minimising

substituting, rearranging or combining the existing features of a product.
Different quality versions of the existing product can be developed so that the needs of
I
different markets can be met.
Additional models and sizes of the existing product can be brought out.
If the company pursues the policy of internal innovation, it implies that it has its o\\r
research and development department which is engaged in the development of new produc>
including modifications and improvement in the existing ones. If, on the other hand, tl;
company pursues contract innovation, it implies that it has engaged the services of outsiu
researchers or new product-development agencies for introducing new products foul
company. Sometimes the company may prefer acquisition while at other times itm
follow the strategy of innovation.
250
New Product Development & Test Marketing
There is a dilemma faced by the management with respect to new product development.
On the one hand, the company finds that it is necessary to develop new products, on the
other, the stake involved in the new product development is very high on account of
research and development activity being highly capital intensive. If the new product fails in
the market, the company has to sustain a heavy loss. It is, therefore, necessary that new
product development be carried out with extreme caution.
New product development is an extremely difficult and time-consuming process. The role
of marketing research in new product development is not as straightforward as it might
appear. Unforeseen situations may arise which may force the company to abandon its
projects mid-way. Such situations arise not infrequently. One has to be fully informed
about the market and product opportunities before venturing into a new product
development project. It is desirable to proceed step-by-step in this process.
13.3 NEW PRODUCT DEVELOPMENT PROCESS

In all the below-mentioned stages, market research techniques can be applied though they
would vary in each stage. In the section that follows, these stages and the major marketing
research techniques used therein are discussed.
1) Idea Generation
8 The objective of this stage is to obtain t new
ideas for products i new attributes for the
existing products v • new uses of the
existing products
There are several sources of new-product ideas such as customers, company salesmen,
dealers, scientists, competitors, the top management, industrial consultants, advertising
agencies, marketing research firms, industrial publications, universities and commerical
laboratories.
Several methods can be used to generate new ideas. These are briefly discussed
below.
251
Marketing Research
Attribute listing
Major attributes of an existing product are listed. Then, one is asked to imagine ho
each of these attributes can be modified so that the product will improve. In t
connection, Osborn suggested that new ideas can be generated with reference tod
particular product. Can it be put to other uses? Can these be
• Adapted ,
• Modified
• Magnified
• Minified
• Substituted
• Rearranged
• Reversed
• Combined
Forced relationships
This technique involves the listing of several objects and then trying to find how e
object can be combined with the other objects. For example, a tyre and tube, f
separate products, were combined into one - tyre and tube - fulfilling a 'felt need1 \
using in a limited space.
Morphological analysis
This term refers to a variety of techniques which are similar to forced relations! and
attribute listing. Although there are several variations, a simple meth
morphological analysis consists of the following stages: First, the parameters oft
situation are listed. Second, each parameter is sub-divided into its smallest nai
Third, these parts are represented in a matrix. Finally, all possible combination^
parameters and their sub-parts are examined. In this way, the morphological a will
enable identification of the components of current successful products and ft new
combinations of attractive features. Such an analysis has been extremely sue* in the
development of new techologies. No doubt morphological analysis MI
consuming. But, the time spent is justified. A thorough search of all
combinations would not be possible without morphological analysis.
252
113 New Product Development & Test Marketing
Problem analysis
Here, the consumers are approached to find out if they have experienced any problem
while using a particular product or product category. One can then select one or two
major problems from such a list on the basis of their importance, the frequency of
their occurrence, and the cost of effecting improvement in the product.
Brainstorming
This technique involves the use of a small number (usually between six and ten) of
consumers who are asked to participate in a 'brainstorming' session. The purpose of
such a session is to generate a number of new product ideas. In order to ensure that
abrainstorming session is most effective, it is necessary to comply with certain rules
suggested by Osborn. These are: (a) No criticism of any new idea should be made,
(b)Freewheeling is welcomed, indicating that the wilder the idea, the better it is.
(c)A good number of ideas must be generated (Quantity is important at this stage)
(d)Participants should suggest how two or more ideas can be combined into still
another idea.
Synectics
Some authors feel that a major limitation of brainstorming session is that it produces
solutions too quickly before developing some perspectives. Instead of defining the
problem specifically as in brainstorming sessions, the Synectics approach suggested
by Gordon defines the problem so broadly that the participants in the group have no
idea of the specific problem, hi such an exercise, the participants give their viewpoints
and as more and more facts are gradually interjected, their discussion tends to move
towards specificity.
Concept Development and Testing

It should be obvious that all the new product ideas generated cannot be pursued.
This may be on account of several reasons. The company may find that a particular
new product idea is incompatible with its major objectives. Further, it may not have
the requisite production or marketing skills. Another possible reason for not pursuing
anew product idea is that it may not be technologically feasible. Thus, a preliminary
screening will eliminate a number of new product ideas. Those which survive this
screening are then pursued further through concept testing. The major objectives of
concept testing are as follows:
253
Marketing Research
Another major difference between the product testing stage and the
development and testing stage is that the former involves the trial use of th
by a group of respondents while the latter attempts to measure only the initial i
in the proposed product. Here, some kind of usage test is undertaken to
whether the respondents would be interested in it and whether they would sut
buy it if it were available in the market.
Usage tests
The new product can be tested in different types of usage situations. There
types of usage tests: laboratory usage tests and consumer usage tests.
In the former test, R and D people may test a new product with respect t
more attributes. For example, a car manufacturing unit will have its R and D c
whose task is to effect improvements in the car, say, economy in fuel consui
and D personnel may evaluate the proposed vehicle by undertaking test driv
varying conditions. Generally, laboratory testing is found to be more rigon
ensuring that performance norms come up to a certain level.
A point worth noting is that a consumer usage test and the laboratory test i
different evaluations of the same product. This is because the consumers' ] m
be different and accordingly they may attach importance to those < whic
were not considered by the R and D department. In a consumer us small
number of consumers are given a sample of the new product. They i to use
in a normal fashion and later indicate their reaction to it as also thel that the
have noticed.
A variant of the consumer usage test is the blind usage test. In such a test, the c
is given the product in question without disclosing its brand name or adve
This method is particularly useful when a new formulation of an establishe
to be tested.
In usage tests, the researcher should ensure that respondents have actually
new product and that their responses are free from bias. This is important £
time respondents indicate their willingness to buy the new product without e
doing so. This would give a highly favourable picture of the product.
Business Analysis
New product ideas that survive the product evaluation and developmert th
taken up for an in-depth analysis to ascertain their business attractive
256
Unit 13 New Product Development & Test Marketing
;ept this purpose, it is necessary to project the future sales, costs and profit, and if such
duct estimates are reasonably good, the product in question is commercialised. However,
as new information becomes available, the estimates of sales, costs and profit may
lout have to be revised.
tently
In order to carry out business analysis and commercialisation of the new product,
two important techniques—test marketing and simulated test marketing—are used.
Both these techniques are based on the application of experimental and
quasi-experimental designs. The increasing use of these designs show that
ire two they are dependable and yield reasonably good results.
When a new product concept has scored high in a business analysis, it is passed on
»one or to the R and D department which develops one or more physical versions of the
lartment accepted product concept. It develops a prototype that satisfies the predetermined
iption.R criteria.
es under
jrous for
13.4 TEST MARKETING __________________________________________
After the concept testing and development of a new product, it is necessary to find out
may give whether it is going to be accepted or not in the market. This is achieved through test
marketing. The main objective of test marketing a new product is to reduce the commercial
attributes
sage test, a risk when it is brought in the market.
y are asked
Test marketing is a controlled strategy done in a limited and careful selected way in a
the defects
selected market place where in the main objective is to predict the sales or profit.
From the above definition, it is clear that test marketing is essentially an exercise in
ne consumer
vertisement. experimentation, where the marketplace is a laboratory. It also brings out that the
shed brand is predictability of sales or profit is the objective of test marketing.
13.5 USES OF TEST MARKETING __________________________________

ually used the
tantasmanya There are two uses of test marketing. First, it may be used as a tool for managerial control.
\outeventuaUy Second, it may also be used as a predictive research tool. These two uses are discussed
msomedetail here.
Test Marketing as a Managerial Control Tool
pment stage are lest marketing frequently serves as a pilot operation for large-scale marketing activity.
tractiveness.For Whenacompany intends to expand its business operations, some element of risk is involved.
25
7
Marketing Research
This is particularly true in respect of new products or brands where the management m
be understandably apprehensive. There are, for example, many physical problems such
the handling of the product, breakage, storage, stocking and transportation which if
handled properly, may prove to be costly mistakes. In such cases, test marketing is u
to improve the mechanics of the marketing operation so that the management may know
advance the problems that are likely to arise and hence improve its marketing operation
Here, the role of test marketing is limited. It is not advisable to undertake an elaborate a
time-consuming test marketing as this may be unnecessary. It is used as a device to g
experience, to identify the problems likely to arise in marketing operations, and to eventua
develop a sound marketing programme. It is not used for purposes of prediction of sal
or profit.
Test Marketing as a Predictive Research Tool

Test marketing is often used as a predictive research tool in two different situations. H
are: (i) the introduction of a new product or brand, and (ii) the evaluation of alterna
marketing variab
Test Marketing for New Products or Brands ^

When a new product or brand is to be introduced in the marketplace, the manageme
apprehensive about its performance. This will be true particularly where the new pro^
does not have any direct substitutes. In such a case, the management neither ha^
information of its own nor any experience of the others. In such a situation, test mark
can be used to predict the likely performance of the new product. W
More often, one finds that test marketing is used for a new brand of an existing pr«
category. However, a new brand may be substantially different from the other^ inB
jR
product category or it may be just another brand and may not be very different fronH
others. In the former case, some form of blind product testing is desirable to fmdB
whether the new brand is more acceptable to the consumers. In the latter case in \«
new brand is just another addition to the product category, test marketing ma\ '«
desirable. When heavy investment is required for the new brand, it may be ad\ ISM
run a test market to ensure its market acceptability first. M
Test marketing is also used while recycling an existing brand. When a company i- xeriB
considering a new packaging for its product, its repositioning, or some improve:nenB
existing brand, it may use test marketing before introducing such a change. In .asJ|
existing brand, test marketing should be used only when the change is substantial otll
258 it will be only a wasteful expenditure. •
When a test market is conducted for a new product or new brand, the management would
like to know how this new product or new brand will do in terms of sales and profits at the
national level. In this regard, three approaches can be used.
1. Buying Income Method: Here the test area sales of the new product/brand are
expanded by the ratio of national income to the test area's income. Thus,
Estimate of national = —-— mcomeo — ecountr y x Test area sales

sales Test area income
2. Sales Ratio Method: Here the test area sales of the new product/brand are expanded
by the ratio of national sales of other product to test area sales of this other product.
Thus,
Estimate of National Sales of other product Test area sales

x
national sales = of test product
Test area sales of this product
This 'other' product or brand should be related to the test product in some way. the
management should have sufficient evidence to think that the relationship in regard to
the 'other' product would hold good with respect to the test product.
3. Share-of-Market Method: Here the ratio of test area sales of new product to the
test area sales of the whole product category is to be multiplied by the national sales
of this whole product category. Thus,
Estimate of Test area sales of new brand National sales of

x
national sales = this whole product
Test area sales of this whole product
Of these three methods, the share-of-the-market method yielded the most accurate results.
At the same time, since it needs information of all the competing brands, it is very costly.
Subsequently, other methods in the form of mathematical models have been suggested for
projecting test market results and are generally based on the consumers' panel data. It is
sported that the use of panel data in such models has given very accurate projections of
sales of new brands at the national level. The discussion of these models is beyond the
scope of this book.
259
Marketing Research
Test Marketing for Evaluating Alternative Marketing Variables
Another predictive use of test marketing is to evaluate the alternative individual marketing
variables. For example, test marketing may be used to find whether a new media pattern
is better than the existing one and, if so, to what extent. Similarly, one may use test marketing
to ascertain whether a higher advertising budget is more profitable than a lower one. One
can think of several situations where test marketing can be applied to measure the effect of
a variable.
However, one has to be extremely careful in designing a test so that it gives dependable
projections. Alittle later we shall discuss design methodology for a test market. Amore
important topic is discussed here: should test marketing be conducted or not?
13.6 GUIDELINES FOR TEST MARKETING

The management is often faced with the dilemma of going in for test marketing or not. This
decision has to be based on a cost-benefit analysis. Against the expected benefits, the
management has to examine various costs involved in a test market. As the likely benefits
of test marketing have been discussed, here the discussion is confined to the other aspect,
namely, costs.
Both direct and indirect costs are involved in test marketing. The direct costs include (i)
cost of setting up a pilot plant (ii) commercials (iii) advertising (iv) expenses incurred on
the production of point-of-sale material (v) couponing (vi) sampling and (vii) trade
allowances offered to obtain distribution. An indirect cost, for example, would be the cost
of disclosing a new product idea to a competitor. Other indirect costs are: (i) deployment
of sales force from the existing and established products to test marketing (ii) opportunity
cost of sales lost in case of a successful introduction (iii) possible negative impact on other
products, etc. The fact that test marketing would mean high direct and indirect costs,
necessitates that there should be a very careful analysis of benefits and costs before using fe
this device. F
It is advisable to spend adequately on a test market so that one is able to get decisive j
results rather than to save on its designing and implementation, reducing its scope and I
thereby obtaining inconclusive results. In the latter case, the entire expenditure on a test | 1.
market will be in vain. I
In deciding whether or not to undertake a test market, major problems associated will F
test marketing must also be considered. An overwhelmingly important problem is the ki F of
projectivity of the test market results. This is due to a number of reasons. One posî;
260
L
r *<-
reason could be that a company's competitors might have learnt about its test marketing
and engage themselves in monitoring its results. They may then come out with a similar
product leading to stiff competition.
When to Test Market

Having looked into both the costs and benefits and the problems likely to arise in test
marketing, the company may like to decide the timing of a test market. Afew considerations
become relevant in deciding as to when test marketing should be undertaken.
First, the cost and risk of failure should be considered against the profit and probability of
success. Aproduct with low costs and low risk of failure may not need testing.
Second, the investment needed in plant both, for test marketing and national introduction,
should be considered. In case the difference in investment is not much, one should prefer
introducing the product to the national level. In contrast, if the difference is considerable,
test marketing should be undertaken first.
third, the possibility and speed with which the competitors are likely to initiate and/or pre-
empt the proposed product should be considered. The faster the competitive firm's response
(and the more likely it is to do so), the stronger the need to avoid test marketing.
Fourth, the effects of a national failure on the trade and consumers should be considered.
One should ascertain whether the company's reputation and its other products are likely
to suffer in the event of a failure in test marketing.
In the light of these factors, a company may decide in favour of or against launching a test
market. If in a given situation only one factor is relevant, then the company may decide on
the basis of only that factor and this process will be much simpler than in other situations
where two or more factors, mentioned above, are involved.
Guidelines for Designing Market Tests

The following guidelines will be helpful in designing a good market test:
The market test must be representative of the whole. This is an extremely important
aspect otherwise the projectivity of the test will be faulty. In order to achieve this, the
sample markets should be randomly dispersed. It may be desirable to stratify the
universe by regions or states first and then choose individual markets within the regions
or states. The sample markets should be normal and should not be dominated by one
industry.
L 261
Marketing Research
2. It is desirable to run the test for at least a year. This is to ensure that a seasonal
pattern, if existing, does not distort the results. This is particularly necessary in the
case of those products that are likely to have a seasonal pattern. Further, the test
market should also reflect a repeat-purchase situation. This is possible only when the
duration of the test market is sufficiently long.
3. The test market must be carefully controlled. As far as possible, extraneous variables
must not be allowed to unduly influence the test. This is particularly relevant in testing
alternative variables where a control group is necessary. Where a control group has
to be set up for comparison with the test area, one should ensure that both should be
selected in an identical manner. Similarly, the control group should be as random!}
dispersed as the test area in addition to being as large as the test area.
4. It is necessary to ensure that the test market gives accurate results. Without a relevant
and accurate measurement of the test market, the projective results will be misleading.
In this connection, three things should be looked into: (i) the measurement criterion
such as sales, market share or profit (ii) the selection of the base period against which
comparisons are to be made (iii) the method of projecting the results of the test
market.
5. It is advisable to test only one variable at a time since a large area is needed to test
even one variable. Moreover, the introduction of another variable in the same test
market may lead to unnecessary complications.
6. In order to ensure that projections made on the basis of a market test are realistic, it
is advisable to use at least one year's back data as the base. This will enable us to
choose the most appropriate method for forecasting. Sometimes, it may be desirable
to use more than one method for forecasting and to average the projections obtained
from these methods.
7. The basic principles in market tests should never be sacrificed for the sakeo
expediency. A market test where several important principles are compromised^
lead to confusing and misleading results.
8. It is desirable to undertake basic marketing research prior to the test marketing.]

Adequate research done before running a test market will give more reliable results, j
At the same time, it may indicate that test marketing is not needed in certain c
Greater attention should, therefore be given to pre-market test research in orderlj
derive the maximum advantage from the test marketing.
262
o4l . .
Other Approaches to Test Marketing
There are some other types of test marketing which are different from what has been
mentioned above. One approach is the controlled test market or mini-market test.
In the controlled test market, the test is run in small area/areas where an effective control
over most of the marketing variables can be exercised. The advantages of the controlled
or mini-market test is that it can be conducted in a very short time and is extremely
economical. One disadvantage is that it does not have projectability. As Achenbaum has
said, "mini-market testing is an excuse for real testing: like so much in research, some are
more concerned with cost and speed irrespective of utility."
The second type of test marketing is known as the pre-emptive roll-out. When a company
feels confident about the new product and finds that competition is going to be heavy, it
may use this method. It may introduce the new product in a few markets or in a particular
region of the country. Although such markets may not be representative of the entire region
or country, they may give some indications of the acceptability of the new product.
Obviously, introducing the product in a few markets would involve much less risk as
compared to that in a national introduction.
Another type of test is known as the simulated test market. High costs of test marketing
have led to the development of simulated test markets, especially for frequently purchased
consumer goods. These tests are conducted in a laboratory environment and involve in-
state shopping. The method is normally used in measuring an alternative marketing variable
rather than a new-product introduction. Here, the stores in a marketplace are split into
two groups, each containing one approach. It is necessary to control test markets and to
ensure that extraneous considerations in the two groups do not distort the test.
13.7 ADVERTISING RESEARCH
A study done by the Centre for Media Studies emphasises the phenomenal growth in
advertising that has taken place in India in recent years. It observes that a number of
factors have contributed to this growth. First, advertising expenditure has crossed the Rs
1000 crore mark. Second, advertising through TV has become a force to reckon with.
Third, consolidation and realignment of advertising agencies has taken place. Fourth,
newspapers and magazines have to woo advertisers and advertising agencies. Fifth,
advertising expenditure is no longer questioned following the new buoyancy in the market.
I 263
Marketing Research
13.8 IMPORTANCE OF ADVERTISING

Another study shows that five consumer megatrends have swept across the urban landscape
in India. These are: the new individualism, the growing multi-culture orientation, the search
for status, a harder look at brand values, and the new aspirations among the consumers.
These megatrends have emerged as a result of four forces of change in the Indian society.
These forces are: more purchasing power with consumers, the availability of a wide variety
of products and brands, the availability of greater information about products and brands,
and changes in the socio-economic environment in the country. In view of these
developments, advertising has assumed more importance in recent years. This rising trend
will continue on account of a number of factors such as the increase in urbanisation, the
expansion of education, the increase in the per capita income, etc.
This would mean that companies will give far more attention to their advertising efforts,
use more sophisticated and improved advertising techniques and be more concerned about
knowing the impact of their advertising efforts.
13.9 NEED FOR ADVERTISING RESEARCH _______________

The role of research in advertising can be seen in various stages of advertising planning.
First, what should be the objectives of advertising? Unless the objectives are clear,
advertising cannot be useful. Research would enable the company to be clear in its objectives of
advertising. Second, advertising research should be used for developing a strategy for
marketing the product in question. Further, the selection of the target audience can be
facilitated by advertising research. In this connection, mere demographic classification will
not be adequate. It may be equally necessary to pay attention to product usage behaviour.
Another aspect where research can be useful is the selection of message that an
advertisement should carry and through which media it should be conveyed. This is a
major area of advertising research. Finally, research has to concern itself with the evaluation of
advertising in order to find out whether the expenditure on advertising has been justified or
not. If not, the reasons should be ascertained so that an improvement in advertising can be
made in the future. In short, research can be instrumental in increasing the efficiency of
advertising as a result of which the pay-off from advertising expenditure will increase,
13.10 MEDIA RESEARCH ________________

I
The main issues in media research are
How to choose amongst media types—television, radio and newspapers?
264
How to decide on a specific insert within a particular type of media, say, television?
In order to decide on these two issues, it is necessary to have some data. The Advertising
Research Foundation (ARF) suggested the following types of data for this purpose:
1. Media vehicle distribution: The circulation number for a magazine or newspaper

or the number of television or radio sets available to carry the advertising.
€
2.Media vehicle audience: The number of people exposed to the media vehicle.
This would be larger than the number in (1) above as more than one person reads the
same newspaper/ magazine or watches the same television set.
i
3. Advertising exposure: The number of people exposed to a specific advertisement
in the media vehicle. This number would be less than the number in (2) above as all
those who are exposed to a newspaper/magazine may not notice a particular
advertisement.
W;
4. Advertising perception: The number of people who perceived the advertisement

in question. This number would be less than that in (3) above. The people may perceive
an advertisement because of several factors such as its large size, use of attractive
colours or its positioning in the media vehicle or on account of the product involved.
5. Advertising communication: The number of people who comprehend specific things

about the advertising. This number would obviously be less than the number of people
who merely perceived.
6. Sales response: The number of people who buy the product in question as a result
of advertising. This number would be far less than that in (5) above.
Itmay be noted that of these six categories, there is an interaction of media and message
in the last four categories. This makes it difficult to obtain the numbers for media alone in
respect of these categories. Accordingly, media vehicle data is generally obtained for the
firsttwo categories, viz. media vehicle distribution and media vehicle audience.
Media Audiences
i research comprises, inter alia, the measurement of the size and break-up of
individual vehicle audiences. We will discuss this under two heads (i) print media and
Bradio and television.
265
Marketing Research
Print Media
Prior to the reports of the Audit Bureau of Circulation (ABC), the measures regarding the
individual vehicle audiences were those which the media themselves claimed. Such measures
were rather inflated as any individual media vehicle would suggest that its circulation is far
and wide. Since the ABC's reports are now available, these inflated measures have ceased.
The ABC compiles its report which gives the size of circulation of a newspaper/magazine
on the basis of certified audits.
i
This information, while being useful, is not sufficient. It is difficult to precisely estimate the
size of audience for a particular publication. The data collected by merely asking respondents
as to whether they have looked at a particular copy would be unreliable. This is because
some respondents may regard reading a particular magazine as a status symbol and hence
they may report an exaggerated readership. Likewise, reading of some magazines may be
regarded as below one's status and hence their readership may be reported to be much
less than in reality.
i
!
Another important aspect in determining the audience size for print media is the extent of
duplication between magazines. This is because the readership of three or more magazines
among respondents is quite common. But the data on readership seldom gives the extent
of duplication. The problem is how to get the size of an "unduplicated" audience. Adetailed
study to determine the duplication among a large number of magazines would obviously
be very expensive, not to mention the time it would involve.
Radio and Television
There are four methods to measure the size of the audience for any radio and tele\
programme. These methods are briefly discussed below:
Coincidental method: First of all, a sample of households having a telephone is
selected This is followed by an inquiry on telephone as to whether a particular programme
on radii1 is being listened or being watched on television. Other information such as the
name of the sponsor and the product being advertised is also collected.
The main advantages of this method are that it is quick and economical. It has som'
limitations though. First, the method has to be confined to only those households whid
own telephones. In a country like India, a large number of households do not have telephone
and hence they have to be excluded. Second, since the enquiry has to be conducted wli a
particular programme is in progress, only a limited number of households can be contact
266
in this short duration. Finally, it is extremely difficult to undertake an enquiry with respect
to late night programmes.
Roster recall: As the name implies, a roster or a list of programmes is used to facilitate
respondents to recall as to which programmes were listened to or watched. Respondents
are contacted personally by interviewers.
This method has some major limitations. First, the responses are dependent upon memory.
Secondly, depending on the status or popularity or otherwise of a particular programme,
respondents may give their replies regardless of whether they have listened to a programme
(or seen it). Third, the method is unable to provide information on a continuing basis.
Finally, it is not possible to estimate duplication in the audience as respondents are
approached for programmes within a short time period. It is possible to estimate the
number of persons who watch both programmes, falling within the same time span on
which respondents are being contacted.
Diary method: As the name implies, this method uses a diary for estimating the number
of persons listening to or watching different programmes. A diary, especially designed for
this purpose, is issued to respondents who have agreed to furnish the desired information.
Each respondent records his radio listening or television viewing along with personal data
such as age and sex in this diary. If respondents accurately record their radio listening or
television viewing, this method would give accurate and complete information, eliminating
the errors that may arise due to memory lapse and interviewer bias. Further, it is cheaper
than other methods involving personal interviewing and recall.
However, in practice, one may find that respondents are not so careful in listing the
programmes listened to or viewed by them. Besides, there is a lack of continuity in the
flow of information as the diary method is unable to provide the estimate of an audience,
ay,minute-by-minute. Apart from this, some respondents in the panel may stop giving the
information sought or move to another address. In such a case, how far the panel will
remain representative of the population is a moot question.
The audimeter: An audimeter is an electronic machine attached to a television set. As

•don as the television set is turned on, the machine records it on the tape. In advanced
countries, this method is frequently used. The audimeter ensures a continuous flow of
information which is not possible in any of the earlier methods. This is its main advantage.
toother advantage of this method is that there is complete objectivity in the information
his collected. Moreover, it is possible to have a cumulative estimate of the audience since
leaudimeter sample will be almost the same from month to month. The method suffers
267
Marketing Research
from some limitations as well. First, turning a set on does not necessarily mean that the
programme in question is being watched. Secondly, the method cannot indicate as to who
is watching a programme.
13.11 COPY TESTING ____________________________________________

Another important area in advertising research is copy testing. The word 'copy' is used to
denote an entire advertisement, including the message, pictures, colours, etc., regardless
of the medium in which the advertisement has appeared.
Methods of copy testing can be divided into two categories, viz. 'before' tests and 'after'
tests. The former category includes all those tests that are used in ascertaining the suitability
or otherwise of an advertisement before it is finally released. Their purpose is to effect
improvements in the copy or advertisement. The latter category includes tests to measure
the effectiveness of an advertisement after it has been formally released. Despite this
distinction, at times the difference between the two types of method gets blurred. This is
because some 'before' methods require that an advertisement should be run in one or two
media.
'Before'methods
A number of pre-test methods are used for copy testing. In this section, we will discuss the
following methods:
consumerjury
. • 1 - :- •
portfolio tests
1 rating
scales
1 physiological
methods
1 dummy advertising vehicles Psych
1 on-the-air tests there;u
i inquiries sente/k
I laboratory testing to find (
them. A
Psycho]
view of;
268 laborati
jsycho/o.
*• '
Consumer jury: In this method, a sizeable number of consumers from the target audience
are shown a set of rough and unfinished advertisements. With respect to these advertisements,
they are asked such questions as: Which copy would you prefer to read? Which one
would induce you to buy the product? Which headline is the most interesting one? Paired
comparisons or ranking may be used by respondents. The assumption in this method is
that at least one of the advertisements shown will be liked by them.
Rating scales: This method involves the use of certain standards against which a copy is
tested. The copy is rated on the basis of scale values. As a result, a numerical score is
obtained. It may be added that weights may be assigned to different factors or items on
the basis of which a copy is to be tested, depending on their relative importance or relevance.
This method is generally used by professional advertising agencies which are able to 'rate'
advertisements without any difficulty.
This method has one major advantage as it provides a list of factors against which a copy
is to be tested. However, there are certain limitations. First, the problem is how weights
are to be assigned to different items. Second, different respondents will rate the items
differently. It is difficult to say who is right in his rating. Third, an overall high score of a
copy does not necessarily mean a superior copy. This is because that copy might have
scored high ratings with respect to several items and low ratings with respect to only a few
items. It is these few items which may be extremely relevant in judging the copy.
Portfolio tests: According to this method, a number of alternative copies that are to be
tested are placed in a portfolio. At times, the copies are placed in the dummy copies of
magazines or newspapers. Respondents are given the portfolio and asked to go through it.
After they have done so, they are asked to recall the copies from memory. Such a recall
may be either unaided or aided. The interviewer may facilitate recall by asking about
specific advertisements. The interviewer may further ask the respondent to recall the
advertisement. The extent of recall will indicate the strength of the copy.
Psychological tests: This method uses a number of psychological techniques to find out
fereactions of respondents to a given advertisement. Techniques such as word association,
sentence completion, depth interviewing and story telling are used by trained psychologists
10find out what respondents see in a given advertisement and the influence that it has on
n. As it is extremely difficult to interpret the information obtained on the basis of I
psychological tests, only trained persons should be appointed to carry out this task. In |
w of this, only a small sample can be used for such tests.
I laboratory testing: This method uses mechanical devices to measure the

respondent's I schological responses to a given advertisement. The commonly used tests
are the galvanic
269
Marketing Research
skin response and the eye movement. As regards the former, a device is used to measure
I
changes in the amount of perspiration in the hands. This may be taken as a measure of
emotional change as a response to an advertisement. However, the test is unable to indicate
whether such an emotional change is favourable or unfavourable to an advertisement. In
the latter test, an eye camera registers the continuous movement of the eye as it reads an
advertisement. However, the results obtained from an eye camera are difficult to interpret,
For example, if the eye was fixed on a certain point, could it be interpreted that the
respondent was interested in the advertisement or that he was confused?
Inquiry tests: Some advertisements may invite several inquiries from the readers about a
given product or service. However, it is questionable whether a large number of inquines
can be regarded as a good yardstick for a successful advertisement.
Inquiry tests can take several forms. One way could be to place the same offer in different
copies in different issues of the same magazine/newspaper. These offers are keyed to a
specific advertising copy. If the number of inquiries linked with a particular copy outnumer
the other, it may be concluded that this copy appeals more to readers. Another variant
could be to give the same offer in different advertising copies that appear in different
newspapers or magazines. This assumes that there are only negligible differences among
different media. However, this may not be the case. Sometimes, the same offer is made
through two pieces of copy. One piece of copy is carried in half the copies of the newspaper
or magazine and the second piece of copy is carried in the remaining half. Inquiries received
are then linked to the two pieces of copy.
Inquiry tests can be developed in the form of controlled experiments to ascertain the
impact of an advertisement copy. However, one has to exercise great care in isolating the
effect of other factors from that of advertising.
Simulated sales tests: These tests expose prospective consumers to different pieces of
copy through point-of-purchase displays or direct mail. Thus, one may select two group
of similar stores where two alternative pieces of copy are displayed at the entrance oral
some other place in the store. Sales of the product in question are measured both beta
and after the display of copy in the two groups of store. The copy in those stores whin
Scores
have registered a higher increase in the sale of the product over time is regarded as;
adverti
better copy. Likewise, comparisons can be made between two pieces of copy using
related
dim; mail.
I
determ
While these tests are both more economical and simpler than actual sales tests, one The rec
niaj« limitation is that there is no certainty that the advertisement when actually
atlveni.s
given wilfc the same result as at the time of the test.
forget h
270
r
i*
Day-after recall tests: These tests are generally undertaken for television commercials.
The test involves an on-air exposure of a commercial in a couple of cities. This is followed
by a telephonic enquiry of the respondents to find out if they can recall the message. The
aggregate recall score that is arrived at is compared with a standard sc ore based on
similar studies. If the score given by the commercial is higher than the standard score, it is
inferred that the advertisement is useful and should be telecast on a larger television network.
The main advantage of this test is that it is performed in a natural setting. Moreover, a
proper sample design can be used in this method. In contrast, the major limitation is that it
turns out to be a test of the respondent's ability to remember. This does not necessarily
establish that the respondent will behave in a different way as a result of watching the
commercial on television. How far can the recall be related to a change in the respondents'
attitude and behaviour? This is a pertinent question which is difficult to answer in the
context of day-after recall tests.
'After'tests
There are three methods that are frequently used to test an advertisement after its formal
release. These are recognition test, recall test and sales test. They are described below.
Recognition tests: These tests are carried out with respect to a printed advertisement
ind commonly referred to as a readership study. Here, the respondents are asked if they
tee read a particular issue of a magazine. They are further asked as to what they saw and
read. Generally, the respondent is shown a particular page of the magazine and then the
iowing measures of recognition are taken:
Noted—the percentage of readers who have seen the advertisement earlier.
fen-Associated—the percentage of readers who read a part of the advertisement which

icates the brand or advertiser.
ad Most—the percentage of readers who read a major part of the advertisement.
Scores are assigned to these three measures and overall scores are determined for all the
1 fatisements contained in a particular issue of the magazine. These scores are then
I itedto the expenditure incurred on the advertisements. In this way, cost ratios can be
Iaermined.
aerecognition method has certain limitations. Some respondents may confuse specific
I fatisements with similar or identical advertisements seen elsewhere. Respondents may
1 k'ethaving seen an advertisement earlier or falsely claim that they have seen it.
271
Marketing Research
Recall tests: In this method, respondents are asked to recall specifics of the advertisement.
In the foreign countries, there are some advertising agencies that offer a post-testing
readership service. To begin with, copies of test magazines are sent to a sample of
respondents who are asked to read them in a normal manner. Telephone interviews are
held on the following day. Respondents are read out a list of advertisements and asked to
identify those which they remember and the extent to which they are able to recall. Thus.
scores are assigned to the ability of the respondent to remember the name of the product.
the underlying message contained in the advertisement and their favourable attitude regarding
the advertisement.
Recall tests, no doubt, go beyond recognition tests but it is difficult to say that recall scores
indicate the desired consumer behaviour. Recall scores may be high and yet there may not
be any perceptible change in the consumer behaviour with respect to the product in question.
Sales tests: This method measures the effect of an advertisement on the sale of the
product. The assumption is that changes in sales are as a result of the advertisement..
However, as there are several factors influencing sales, one has to be extremely careful in,
establishing a relationship between advertising and sales. It is desirable to isolate the influence of
other factors while determining the impact of an advertisement on the sale of the product
Experimental studies can be designed to study the impact of an advertisement on sale, j
13.12 EXAMPLES OF ADVERTISING RESEARCH STUDIES IN INDIA
There are quiet few examples of advertising research done in India. Some of them;
listed below.
Take the example ofLifeboy. The soap was introduced for the blue collar segments
the caption tandurusti ki raksha kare Lifeboy this was for a particular segment afterA
re-launching of the product now it comes in Orange, Blue and White wrappers. This is!
the entire family.
In the above example, the segment for only a specific class, as it did not have much of«
sales the product was re-launched as "Family Surakshaa" which is doing good™
market. 1
Second example is of Aritel. We remember the mushy grandparents and long last gran
reunion that the airtel ad released sometimes ago. Now the new ad is sequel the gran
is on his way back to the city leaving behind a visibly anguished grandfather in hisa
who even planned a game of chess that evening to cheer him up. The grandson cii
up from the train to play his move on the chess board. 1
The message is single minded to underscore Airtel's vast connectivity.
Another example of Fair & Lovely if you remember, it showed Mr. Kale, Mrs. Kale, and
Miss.... He wanted to say Miss Kale but the girl was very fair though the father and the
mother are showed dark, the girl is fair. The ad is in Marathi and is aired only in the state
of Maharashtra. Now Kale in Marathi means Black and Gore in Marathi means fair.
Now they are showing Khoobsurati ki shakti the ad says that you only need to be fair
and have little "Jhatak Matak" to become a film star. They have taken into consideration
everybody dreams of becoming a film star.
Not only the company changes there ad on the basis of the research but the rival competitors
also copy the punch line of the ad.
Amir Khan's ad of burp burp of Coco Cola, well Dabur has decided to go with slurp,
slurp with their Real oranges.
The ad of surf Excel Dag Achha hai bunch of little boys playing cricket, they go through
the motions of the Indian cricket team expressing leaps and celebrations all through and
get nicely muddied and dirty.
While now they are showing a little boy playing, his grandmother is weaving a sweater
suddenly the wollen ball falls down from the grandmas' lap, the little boy runs after it falls
down, falls in the mud and brings back the ball to his grandma and says "smile please".
One may observe that they have not changed their target customer but have given more
weightage to the punch line dag achae hai.
In the cosmetic sections if we take the ads of Charmis, Emami, Nivea, Lakme and
Ponds each ad has unique attributes.
Now what they want to show of their product, if we consider the example of Nivea their
lengths can be listed as
i High quality product

i Expensive • Widely
available
i Complete range of skin care products i
International brand Suitable for Indian Skin
273
Marketing Research
• Expert on skin care
But if consider Lakme then the attributes would be
• For younger women

• Widely available
• Popular brand
• Offer high quality product
• For Fashionable women
While in ponds it is some what

.
'
I
• For middle class women (working Class)
• Widely available
• Popular brand
• Good value for money
• Used by older women
Now, from where these attributes have been generated? It is from the research and the
impact of the ads that have been aired.
The question comes to our mind why ad research is necessary? The impact of the ads
have to be on the target customer and the segment which they are looking for if the message
is portrayed clearly. The sales volumes increases hence monitoring of ads is a key factor
for all FMCG products because these products sell on ads and promotions only.
In the past, marketers have often relied upon the strengths of their brands for positioning
platform. Now the crucial factor of the importance is the attributes or the parameters 01
the strengths which affects the purchasing decisions. The criticality of this factor cannot be
emphasised, especially in the context of sustained application of the mind churning out the
attributes and parameters, some of which at least will have only a tangential relevance ti
the issue of purchasing decision.
Now how does these help if we consider the example of Nivea. It cues the \
'expert of skin care' this plays an very important role if some body does not prefer Ni\
-it was earlier said 'good value for money' and 'complete range of skin products
important cues of "expert of Skin care" drives the preferences for the non preferred
buy Nivea.
T7A
Among the preferrers, the cue position it was seen that 'complete range of Skin products'
and 'offer high quality product' are also important drivers of preferences.
These are been observed in the ad impact survey. While going through these examples
one is clear about how ad research plays an important role in increasing the sales volume.
/& Activity A: I
1. In your day to day life you must have come across many advertising research names,
write any two of your own choice.
13.13 EVALUATION OF ADVERTISING RESEARCH

In the preceding pages, we have seen that a number of measures are used in measuring the
i and the effectiveness of an advertisement/advertising campaign. Both, short-run and long-term
measures are used for this purpose. In the former category are awareness, day-after
recall, gross rating points per rupee, recognition, comprehension and persuasion scores.
3fthe ads e In the latter category, there are changes in customer attitudes and preferences, in brand
message key loyalty, and in sales and market share performance. In this connection, it may be noted
factor ily. that there are certain contradictory statements in the evaluation of advertising research.
positioning While many companies have conceded that short-run measures are not good indicators of
rarneters or ar the long-term impact of an advertising campaign, they seem to devote considerable attention
cannot be tosuch measures.
mingoutthe
relevance to vertising is considered a causative factor influencing market share and sales, it
ige to find that whenever the market share and sales decline, the first casualty is the
advertisingbudget.
the position
prefer Nivea, Marge number of client firms do not seem to have any specified objectives of advertising,
products' the yet advertising agencies seem to tolerate this. These agencies, on the contrary, should
i preferrers to Encourage their client firms to develop specific objectives and also objectives for their
Muse.
275
Marketing Research
13.14 SUMMARY
We have seen the new product development, why it is done, why it should be implemented
when a product goes through a maturity stage. New development is required as people
always want something new. In advertisement though, it seems to be a very simple term
but proper research and a proper media is to be chosen so that the message is been read
by millions of people. The most important thing is, it should be targeted to the target
customers.
13.15 KEYWORDS
Morphological: the study of form of things especially of animals and plants and of words
and their structure
Mega-trends: excellent direction/tendency
13.16 SELF-ASSESSMENT QUESTIONS ____________________________
Ql. List the steps in the new product development.
Q2. Why is advertising research necessary?

s
Q3. Identify the major areas where advertising research can be helpful.
Q4. What are the two main issues involved in media research? Identify the type of data
needed to conduct media research.
Q5. How would you measure the size of television audiences? Evaluate each of the
methods that can be used. ;
Q6. You are the manager of a large departmental store in Pune city. You are interested in
measuring the impact of the public address system on the sale of soft drinks, etc.
Describe how you would go about it.
276
Marketing Research
14.1 INTRODUCTION
" If your report doesn't make the information easy to understand, believable and
actionable.... it is likely that the management won't value your work."
- Nancy Bunn Director, Corporate Communication, Burke Inc.
This statement sums up the importance of proper presentation of research data. No matter
what quality of research undertaken, the acceptance of result depends on the way they are
j
communicated to the target audience. A report serves these functions. i
;
f
1. It is a systematic record of research whereby data analysis and it findings are placed
in an organised and permanent form for future reference.
2. The quality of research is adjudged by the report.
3. Properly organised, lucid reports lead to appropriate actionable policies and decisions.
4. Management perceived the value of marketing research, as atool for repeat use will
depend on the usefulness of the report.
Report writing is both an art and science. Each report is tailor made to adapt to the
character of the problem, information available and the mode / preferences of the people
utilising them. They also reflect on the personality, experience, knowledge, creativity and
training of the person writing it.
Interpretation means explanation or finding out the meaning. It involves drawing inferences
from the analysis of data. Interpretation and analysis are closely interlinked. Analysis of
data often includes a simultaneous interpretation of the results. For example, when the
researcher calculates a coefficient of correlation, he concludes not only on the presence or
otherwise of a relationship between two or more variables but also whether this relationship.
if existing, is statistically significant. Further, he will also be interested to know about the
implications of such an inference in so far as that particular marketing activity/problem is
concerned.
Interpreting data in the proper perspective is very important. In order to do so, the
researcher should possess a high degree of skill and exercise, utmost care and objectivity
For errors of interpretation, if committed, can nullify even the best research.
To begin with, the researcher must ensure that the data collected are reliable and adequa
for drawing interferences. If the data suffers from inadequacies, then even the best of the
280
methods used in analysis and interpretation would be rendered useless.
Unit 14 Presentation of Research Findings
Any research that is based on sampling is naturally subject to a sampling error. In a sample
survey, there is a general tendency on the part of the researcher to jump to conclusions or
generalisations on the basis of too small a sample. For example, survey results based on a
sample of 20 to 30 households cannot be regarded as a representative of the household
population and applied to a city having a population of two lakhs. In addition, there may
be non-sampling errors such as interviewer bias, defective instruments for data collection,
etc. These aspects have been discussed in this book.
Further, one should ensure that the appropriate statistical methods have been used in
analysing the data. One can use univariate, bivariate and multivariate analysis and several
methods within the last two categories. It is advisable to take the help of the experts in
selecting one or more appropriate methods of data analysis. A frequent source of
interpretative inadequacy is the neglect of measurement problems. The marketing research
must ensure that the measures used are reliable and valid. Absence of this check may lead
to a serious mistake.
Afallacy in the interpretation of the data arises when the researcher tfiinks that all differences
in numbers are really meaningful. This is the area of statistical significance, a concept which
is not understood by a large number of readers. Particularly in those cases where the
difference in numbers is not large, the researcher must indicate whether this is statistically
significant or not. In the absence of such a statement, inconsequential differences will be
interpreted as real differences thereby leading to erroneous conclusions.
Further, negative or inconclusive results are more difficult to interpret than positive ones.
When the results are positive and are in conformity with an established theory, it is far
easierto interpret them. However, when they are negative, then they may be so on account
of the wrong methodology, inadequate or poor measurement, faulty analysis and incorrect
hypothesis or theory. In such a situation, one has to scrutinize the negative results very
carefully to identify the real cause. If one is satisfied with the methodology, the measurement
asalso the analysis, then one can be confident that the hypothesis or theory is not correct.
This in itself may be a definite contribution to scientific advance.
Finally, it must be remembered that the data analysed related to a single point of time in the
past. As such, the situation might have undergone a change and one must not be rigid in
adhering to the same results where the situation is a dynamic one.
tee are some of the major flaws that might occur in the interpretation of data. The
problems specified above are merely illustrative. The marketing researcher should take
are to avoid such mistakes otherwise he will reach the wrong conclusions. Having
mphasised the need for the proper interpretation of the data, we now turn to report
281
Marketing Research
writing. The rest of this unit provides broad guidelines for the preparation of research
reports.
There are various types of reports. But before we discuss them in detail, let us understand
how one should write a report.
« i *j
.1^
Report Writing
I
After the data has been analysed and the conclusions reached, the marketing researcher
has to report his findings to the management. The reporting has to be done in a clear
manner so that the chances of any misunderstanding can be minimised.
Oral and Written Reports
Differences between oral and written reports are as follows:
• First, oral reports lack visual advantages. No charts, diagrams or pictures can be
used, though, to stress particular points in an oral presentation, the reporter can use
pauses and volume emphasis.
• Second, the pace of presentation cannot be controlled and regulated by anyone from
the audience who is being presented an oral report. In contrast, whenever certain
point is not clear in a written report, the reader can read it two or more times, if need
be slowly and more carefully. This advantage is not available to one who is receiving
the report through an oral presentation.
• Third, since a written report is bound to receive considerable attention and scrutiny
from the readers, the report-writer is likely to work hard to produce an accurate
report of a high quality. In contrast, an oral report will not be so precise nor will the
reporter give as much time in its presentation since it cannot be subjected to the same
degree of scrutiny as written reports because of the extremely limited time at the
disposal of the audience for indicating its reaction. Also, established standards for
oral presentation are far less rigid than for written reports.
Oral Reporting -A m
If the researcher has been asked to make an oral presentation of his research findings, he
should bear in mind a few major considerations.
First, he should know the audience to whom he is to report. If it has any special
characteristics such as being highly qualified or specialising in a certain subject, these must
be noted down.
282
Another consideration in oral reporting is that it should be properly planned. To begin

with, the reporter should be clear about the objectives of the report. Does he want to
inform the audience of his results? Does he want to persuade the listeners to agree with his
viewpoint? Does he want to recommend a certain course of action based on his study? In
each of these cases, the oral presentation will be different.
After the reporter has decided the objective of his report, he should gather the necessary
information. If he has not already done any research, he may be required to uniformly
collect data. He may have to approach some people to get the requisite information or he
may collect it on the basis of his own observations.
Once the information is ready, the reporter has to organise it in a logical manner. The
subject matter should be divided into meaningful and comparable parts. Simple ideas
should precede complex and difficult ones. There should be a logical and coherent approach
in presenting the subject matter before an audience.
Another point worth considering is the suitability of language. The reporting should be
done in a simple and convincing manner. In order to make the communication effective, it
may be advisable for the reporter to use visual aids such as charts, diagrams, tables,
pictures, posters, blackboards, slides and movies. However, excessive use of visual aids
may sometimes spoil the overall quality of presentation, therefore, it is necessary to ensure
their judicious use.
Sometimes, it may be advisable to distribute a handout containing statistical data or charts

to the audience. This would facilitate both the listener and the reporter. The latter can refer
to the figures or charts in the handout as he proceeds with his reporting.
When an oral summary of a written report is to be presented, the reporter should exercise
great care in the preparation of the summary. All major points must be covered and
excessive details avoided, otherwise the focus will be lost. The reporter should ensure that
within the time allotted to him, he covers all major points, including his recommendations.
Written Reports
Types of reports
Various authors have classified reports differently. These classifications indicate the variation
in their approaches to the subject. One basis of classification is the time interval. The
reports could be daily, weekly, monthly, quarterly, or annual. However, all reports are not
regularly brought out. There are some which are prepared for a special assignment and
are known as special reports.
283
Marketing Research
Reports are also classified on a functional basis, according to which there could be three
types of reports-informational, examination, and analytical (also known as problem-solving).
The informational report is a factual report on a particular subject and it neither contains an
analysis nor conclusions. The examination report goes a step further. In addition to giving
facts on a subject-matter, it analyses these facts. It does not specify conclusions or
recommendations though the reader can draw these on the basis of the factual information
and on the analysis contained in the report. The analytical report goes a step further than
the examination report. It not only presents and analyses data but it also draws conclusions
and, if necessary, makes recommendations. This is, obviously, the most comprehensive
report.
Reports can be classified on the basis of their subject matter or field. Since there are
innumerable subjects, the types of reports within a certain subject will be different from
those in another. For example, reports on economics could be classified by various subjects
covered within its scope such as a report on monetary reforms.
Reports can be classified on the basis of their physical form or make-up. The two usual
forms are short-form and long-form. For example, memoranda, letters and progress reports
may follow a short-form, whereas information and examination reports may follow a
long-form.
Reports have also been classified on the basis of the relationship between the reader and
writer. Such reports are described as administrative, professional and independent. An
administrative report is written within an organisation, while a professional report is
submitted by an outside expert to an organisation. Independent reports are generally
prepared by non-profit research organisations and published for the benefit of the public.
These reports are not written for any particular group.
Reports are sometimes classified on the basis of the employment status of their authors
Reports prepared in a business organisation are known as private reports while those
prepared in a government or public institution are known as public reports. Here, too.
, there may be independent reports which are prepared by individuals without any
' authorisation either by a public or private agency.
i
Yet another basis of classification of reports is their formality. On this basis, the reports j
may be either formal or informal. The dividing line between the two types is rather hazy. A |
formal report is written befitting the requirements of a formal occasion. An informal report. ,
on the other hand, may be worded to suit an informal occasion. A report prepared by one i
employee for the information of his colleagues may be an example of an informal report. 1
284
contrast, a report prepared by an executive for the top management may be a formal
report.
These are the maj or bases of the classification of reports though it may be noted that these
are not mutually exclusive. Thus, a research report may be a special report, a long report,
an analytical report and a private report at one and the same time. This indicates that there
are different ways-which are not mutually exclusive of classifying reports on account of the
different viewpoints or approaches involved.
Preparation of the Report
Having decided on the type of report, the report writer should now concern himself with
its preparation. This can best be done when he is clear about what aspects or points are to
be covered by it. Let us first consider the format of the research report.
Research Report Format
Aresearch report can be written in a number of ways. However, three formats are generally
followed. First, the report may use a logical pattern, which implies that the findings are
presented in the inductive order, i.e. moving from specific to general. Second, the report
may follow a psychological pattern which is almost inverse of the preceding pattern. In this
format, the most critical information, i.e. the conclusion, is provided first, after which follow
the findings supporting the conclusion. Thkd, the report may use a chronological format
wherein information is given along the time dimension, i.e. the things which happened
earlier precede those which happened later. This form is generally combined with other
formats. The chronological format is the least popular though in respect of the problems of
ahistorical nature, it is the most appropriate one.
Having decided the format to be used in a research report, the researcher has now to
prepare the report outline in accordance with that format.
Report Outline
Before attempting to write any report, the researcher must prepare a report outline. Without
any outline, his report is bound to be haphazard and clumsy. An outline is to the writer that
the blueprint is to the construction engineer or what the pattern is to the dressmaker. The
outline will not only guide the writer as to the order of presentation of ideas but will also
enable him to think before writing. This will lead to clarity in his presentation.
While preparing an outline, it must be noted that it should not be restrictive and rigid,
There should be flexibility built into it so that if a change is subsequently needed, it can be
285
Marketing Research
easily introduced. Too frequent changes should be avoided. If they seem to be necessary,
they will only indicate that the task of outlining the report was not done properly. It should
also be noted that a change of one point in the outline may involve one or more related
changes elsewhere, for that point may have a bearing on one or more other points contained
in the outline.
14.2 TYPES OF REPORTS ________ , _________
A report is composed of working papers and preliminary draft meant for the use of the
researcher and a basis for the preparation of the final draft.
Report for Publication
These reports vary in content, style and character according to the magazine, journal or
bulletin or the interest of the target audience. These are generally concise formats without
much verbosity.
•"'I
Technical Reports
These reports are meant for scientific, or technically trained people. They give minute
details of planning data collection, analysis and interpretation, including bibliography and
secondary sources for substantiation and the cross-examination of results.
I*
!
Executive Reports
:
These reports are meant for decision makers who are generally busy. Voluminous detaiK
are avoided in such reports or put in an appendix. Key findings, assumptions, conclusn
and recommendations are stated in the reports. î
14.3 REPORT CONTENT ___________________ f
It is difficult to devise a standard pattern / format for report writing. Given below is a useiul
outline of report for executives:
1. Letter of transmission
2. Title Page
3. Table of Contents
4. Executive Summary
286 5. Introduction
6. Methodology >
7. Findings
8. Limitations
9. Conclusions & Recommendations
10. Appendix , .»-..
11. Glossary & Bibliography
This is a conventional form and the logical sequencing in report writing.
1. Letter of Transmission: Indicates to whom the report is submitted, reasons for

„ research and an official authorisation for doing the same.
.•48
2. Title Page: Ordinarily, this page is first in the report but some people prefer to put
the transmittal letter first. The Title Page shows the subject of the report, for whom it
was prepared, who prepared it, date of completion and submission. This page creates
the first impression of a person as far as the report in concerned. It should be simple
I and dignified.
3. Table of Contents: It should list all the topics covered, with the appropriate page
numbers. Major heading and sub-headings must be included. The table of contents is
iri followed by list of tables, charts, exhibits, appendices, bibliography, etc. as
a continuation of the same page or on individual page(s).
4. Executive Summary: Is a miniature report which includes major facts, findings and
conclusions in a concise form for a busy executive. It is sometimes the only part that
is read. Hence, it should be made with the utmost care after the entire report has
been written.
5. Introduction: This serves as an orientation to the reader about the problem, nature
and scope of research, objectives to be achieved, historical data and the hypothesis
on which the study is based.
6. Methodology: It deals with the research design, the technique employed, the
justification, sampling design, questionnaire and the unique problems of handling
i experiences.
7. Findings: This is the longest part of the report and comprises several chapters on
details such as general findings, specific market segments, geographical area, etc.
287
Marketing Research
These findings should be organised in a coherent and logical way. Various analytical
and statistical tools help explain the significance of the raw data. Tables, graphs, etc.
can be used to show the relationship of the data analysed. A portion of the detailed
information can be put into the appendix for people who want to study it. This is a
step towards the understanding of the conclusions and the recommendations.
8. Limitations: This provides the readers with the insight into the special conditions
pertaining to the work. Limitations could be caused by time, budget, organisational
and research design constraints. This guides the user as to what degree of generalisation
should be done. This section should be written with balance so that the confidence in
the report is not shattered or undue importance be given to it.
9. Conclusions and Recommendations: Researchers interpret the results in light of

the problem addressed to arrive at conclusions. Based on the results and conclusions
and if the researcher knows the bigger picture of the clients' firm, he/she may make
recommendations. If recommendations are made, they should be feasible, practical
and actionable for direct policy making or any managerial decision.
10. Appendix: It provides materials supplementary to those given in the main report.
For example, the questionnaire used, tables, diagrams, sample data, interviews.
instructions, etc.
11. Glossary: List of technical words and their meanings.
12. Bibliography: It is the final section of the book and contains information of references
and secondary data. For example, names of books, journals, and periodicals.
14.4 PRINCIPLES OF REPORT WRITING__________________________

Report is the end product of research process. It is the means of communication between
the researcher and the decision makers. A report should be written for specific user(s).
keeping in view their interests and the usage of the report. Skills should be used to make
it interesting, understandable and meaningful. Words, figures, illustrations and symbols
must be used for effective report presentation.
Some of the pointers of effective report writing are:

1. Easy to Follow: There should be a logical flow of material. Heading, sub-headings
should be used. Paragraphs should be short and convey one central point in each.
2. Clarity: Clarity in writing should be there lo avoid vagueness and misinterpretation,
288
3. Use good sentence structure: Sentences should be short, grammatically correct

and well constructed. Long sentences are difficult to read and comprehend.
4. Technical Jargon: It should be used when unavoidable. No executive has the patience
of sitting up with a dictionary on his table while working.
Terse: Areport should be terse and concise. It is often said that "Brevity is the soul of wit".
Unnecessary and lengthy explanations should be avoided. However brevity at the expense
of completeness is not intended.
Reinforce text with tables and graphs: Visual aids can greatly facilitate communication
and add to the clarity and impact of the report.
Objective: Objectivity is the true virtue that should guide report writing. Researchers
j must courageously present the facts 'as they are' and not try to colour them to the likings of
the management. Understanding of the objectives provides the framework for presentation.
Presentable and professional appearance: Reports must be professionally reproduced

with quality paper, typing and binding. Variation in type sizes and the skillful use of white
space may attract attention to major and minor parts of the report and thus enhance the
readability and appearance.
14.5 VARIOUS GRAPHIC REPRESENTATIONS AND ETHICAL ISSUES
1. Geographic Maps: Communicate the relative location.
289
Marketing Research
•und chart divided into sections denoting Percentages

2. Pie Chart: It is a
associated
roi
with each value.
• East
• West
D North
IstQtr 2ndQtr 3rd Qtr

4th Qtr
3. Line Chart: A chart that connects a series of data points using continuous lines,
Qtr 2nd Qtr 3rd Qtr 4th Qtr

1st
.'$!*'
4. Pictographs: It is a graphical depiction that uses pictures or symbols to display data.
2000
I
1999
1998
5, Bar Chart: Achart that displays data in bars positioned horizontally or vertically.
4th Qtr
3rd Qtr
2nd Qtr
1st Qtr
20 40 60 80 100
00
1
Marketing Research
6. Histogram: A vertical bar chart in which the height of the bars represent the
cumulative / relative frequency of occurrence.
DEast I
West
(North
1st Qtr 2nd
Qtr 3rd Qtr
4th Qtr
Some Ethical Issues

Report preparation and
presentation
involves issues pertaining
to research integrity.
*
r
1. There should be no deliberate misrepresentations to support personal or corporal;,

views. Presentations should be honest, sober (less dramatic) and done with a neutral j
angle.
i
2. Market problems should not be designed to suit hidden agenda. •
'S
3. Research design should not be compromised by deliberately misusing statistics if

falsifying figures to alter results. ,,,J
4. The researcher should remain objective through all phases of research.
5. Complete confidentiality of respondents and secrecy of data should be maintainei|

(Not to be used for other clients and purposes). m
6. Clients should not distort research findings to develop biased advertising c

' Unit 14 Presentation of Research Findings
£1 Activity A;
a) Give any five points which should be remembered while writing a report.
i
,*.,..
What are some key pointers to keep in mind while writing a report for executives?
Mention any three which you feel are important.
14,6 MANAGING MARKET RESEARCH _______________ _ _ ^ _ _ _ _

j Agood relationship between the client and the researcher is vital to successful research. i
The set of concerns and responsibilities for each should be well defined for effective
management research.
, Client/Researcher Interface
Job Client Researcher Client and

Responsibility Researcher Joint
Responsibility
Problem Monitor competitive
Determination and market environment
Problem area/
opportunity area
determination Research
need determination
Problem definition
293
Marketing Research
Planning of Study of hypotheses Data needs Scope of project

Research Or Alternative actions Analytical Cost and time v/s
Research design
design
Techniques
(Though client
approval must
be taken)
Management concerns: The client may have an inhouse research department. The person
who can initiate a research/maximum user in an organisation may vary. It may be the CEO
of the organisation or a Product Manager. However, there are a few questions which need
to be answered by the client/organisation themselves.
I
1. List of major problems to be studied. ^
2. Does the identified problem need research?
3. Is the study design relevant to objectives?
4. Will the conclusions be valid and applicable?
5. Affordable cost and time lag.
6. Cost v/s benefit of research.
7. Comparison of defined research problem with other problems that need to be studied
8. Inhouse v/s externally sourced research.
Managing the research department: The research director must oversee the execution
of research project at each stage. In case the internal manpower is inadequate or too
inexperienced to handle the project, external help may be sought to do the research.A
ill number of research agencies must be explored to make the right choice. Steps towards
making the right decisions:
Define the exact need for study
Review the services of various firms
I
294
Previous experience with a listed firm (if any)
Check on a research firm's references and clients
Preliminary discussion with seemingly qualifying firms
Proposal invitation with qualifying firms
Competence of proposals
Comparison of fees/ costings
Ascertain the actual people deputed for work
i> Final selection of the firm
After the final selection of the firm, a detailed contract should be drawn to assure a good
performance by the research agency and no shortcuts should be attempted.
The written contract should cover the following aspects:

i • ' ' " ' "
| 1. Quality control for the validation of the data 1
Subcontracting of part(s) of the work
3, Methodology of research
4. Formats of report/presentations
x Right of report publication
3. Penalty clause covering deficits such as delays, inadequate data I
Amount, basis and timing of payments (fees)
8. Client servicing designates
9, Schedule of work progress
I 10, Progress review meetings schedule ,. Feedback on
report findings and implementation
295
Marketing Research
The client and researcher relationship should be of trust and warmth to ensure a smooth
working. However, it should not become too personal to allow laxity in work.
Resources for Marketing Research
1. Budgetary allocations: Two aspects that need to be answered regarding

finances are:
a) What amount should be allocated for research?
b) Who allocates the amount for each project?
Allocation of funds by the marketing research head is the simplest way of resolving
this issue.
2. Staff: It is people who manage data. Data cannot be maintained without an

experienced and educated research staff. Different jobs require different specialisations
to deal adequately with the situation. Also a high level of self-motivation leads to the
successful and timely completion of research projects.
14.7 SUMMARY __________ •'' ____________________________

We have seen the different topics while writing a report. Though survey plays an important
role, the entire study depends upon the outcome of the report. Doctors also ask to test
some tests. On receiving the report, they get some confidence in their diagnosis. On the
basis of the report, the doctors prescribe medicine. Thus, on the basis of the report, the
top management can take the correct strategies and hence report writing is the heart and
the soul of any study.
••ft
14.8 KEYWORDS ______ • ___________________________

Ethical: Relating to morals

Q1. If you are given a task to write the report on, what would you cover in the introduction
of your report?
Q2. What would you cover in the methodological section?
296
Marketing Research
15.1 INTRODUCTION
In this unit, we are going to learn the modern approach of survey which is carried through
internet media. Services have become a key factor in the modernisation era. Service is
always tangible. One person may like the service, while in the same aspect, another person
may be dissatisfied. Hence, the service organisation is always on the vigil on how to improve
the service or how they can provide the best services to the customers, whether they are
a bank, a hotel, an airline or railways, etc.
15.2 RESEARCH THROUGH INTERNET ERA _________________
In the present age of connectivity on the internet, it is possible to do research through

e-mail or through a website which can be accessed by respondents. The biggest
advantage is the speed of response. With proper enabled software, the entire
process can be automated. Both, text and graphic form can be shown. Interactivity with
respondents is possible with net-based applications.
Research in Hotel Industry
When we think about hotels, they come under the service industry, just like the airlines,
railways, banks, etc. As we know, services cannot be measured. If one is happy with the j.
service, it is his own experience. It may not be the same with the other person, he may feel '
that he has not been given the proper service. Hence, it all depends upon our experience [
The service industry, which is very demanding in this era, has also got tools to rank their f
services. Hence, they also do the survey as per their requirements. Let us take an example j
of a hotel industry service. What are the service parameters they measure? Given are (
some of the parameters which are usually measured:
• Did you avail of the airport pick up facility?
• Were our attendants present at the airport when you checked in?
• Did they carry placards with your family name written on them?
• Was your booking confirmed when you checked in?
• How was the reception counter?
• How was the ambience of the hotel?
• Had your room been cleaned when you entered it?
300
Unit 15 Research in Service Segments
• Had all the linen, blankets, pillows, bath towels, toilets been cleaned when you checked
in your room?
[through
ervice is » Were all the electronic equipments like TV, phones, air conditioner, etc. in the
er person working condition?
(improve
r they are « Was the entire toiletry like soap, shampoo, shower cap, combs, etc. available?
• Was the confectionary and mini bar fully equipped with soft drinks, biscuits, chocolates,
mineral water?
» How was the room service?

hrough
e-idvanta » How was the taste of the food which you ordered?
ge ss can
be These are some of the questions which have been asked to the customers visiting the
ondents is hotel. On the basis of these questionnaire, the parameters are ranked and the necessary
actions/improvements are done. In this unit you will find a sample questionnaire on the
bank's credit card services about the customer satisfaction survey, which will give you an
idea about how a survey is done in the service industry.
ie airlines,
jy with the 15.3 METHODS OF ONLINE RESEARCH/SURVEYS ___________________
ie may feel
On line research: Basically, there are three methods of doing online surveys:
;xperience.
Drank their 1. e-mail survey
an example
> Given are 2. HTMLsurvey
3. Downloadable interactive surveys
(•mail: It is the fastest and simplest route out of the three methods. There is little set up
time required and the reach is wide. Anybody having a net address can be sent an e-mail.
This is restricted in simple text messages. Many internet service providers have the list of
segmented address according to socio-geographic and specific interest segments.
HTML: This format uses graphics, grids, sound, sequencing and skip patterns. Setup
time is required to make the format. This format can be used for complex questionnaire.
The HTML format can be mailed to an address,or a simple mail can be sent to the
address to visit a parTtctriarêBsTteafl^'answer I
luestiprmaire. A connecting link is
providedTnsuch casesltflfieend of the e-mail to the referred site.
30
1
• Since all the people do not have net at their homes, the number of respondents is very
low.
ife ETHICAL ISSUES IN MARKETING RESEARCH ________________

Over the past decades, the use and impact of marketing research has been of tremendous
growth. It is agreed that marketing research has great influence on marketing and social
systems. With the rapid rise of research as a form of market intelligence, the researchers
are forced to scrutinise the ethical aspects of their working.
0
Some possible ethical dilemmas are as follows:

1. Undertaking research for direct competitors
2. Different pricing for undertaking similar studies
3. Using excerpts/part of report of one client for another client
4. Giving gifts/benefits to solicit business
5. Claiming authorship of a report/survey which one has not conducted
6. Client distorting the research findings for advertising claims
i 7. Research firms seek information from the secretaries of a rival firm
Ethics in research underline three basic responsibilities are as follows: i

Responsibility towards the respondent
• Responsibility towards the client
i Clients' responsibility towards the researcher
Responsibility towards the Respondents of the Study and Client
1) The purpose of the study should not be to sell merchandise.
2) The anonymity of the respondent, if agreed, must be protected. Various unethical

practices as agreed are stated below and should not be undertaken:
i Ultra violet ink - disguised customer coding
30
3
Marketing Research
• Hidden tape recorders - to record indepth interviews
• One way mirrors - to view customers without their knowledge
• Fake long distance calls - to get a response
• Fake research firm identities - to gain acceptance
These disapproved methods are, however, used by many to make the research task easier.
Right to choose: A person should have the right to participate in the survey or to refuse
participation.
Right to safety: Right to safety means protection from physical harm and psychological
stress.
Right to information: The subjects should have the right to information about the research
proceedings, findings and follow-ups.
Right to privacy: Respondents do not want to be disturbed about partaking in the study
and to express feelings or impart information which is of a personal nature. For example,
unsolicited emails.
«
Abuse of respondents: Abuse of respondents can happen in many ways. For example,
frequent interviewing of females below 30 year, unnecessarily prolonging the interview |
which the subject does not warrant. All such practices are unethical.
15.6 RESPONSIBILITY TOWARDS A CLIENT _______________________

I
Confidentiality: The relationship should be like a doctor with his patient or a lawyer with I
his client.
Protection against unqualified researchers: If the research area and techniques are
beyond the firm's expertise, the study should not be undertaken as it increases time and
cost delays, besides the inaccurate reporting.
Protection against unnecessary research: A client should not be subjected to research

tk&i&nflUQQUAJsd^fQi economic reasons,or that which is already done. *A
304
Protection against abuse of position: A researcher is an expert in his field. Hi> |
recommendations are taken on face value with confidence owingtohis position. Heshouki '
«
not recommend expensive methodology when inexpensive methods can beget adequate
information.
Anonymity: The firm sponsoring the study should not be revealed to any outside party
unless the client so agrees.
Protection of confidential and proprietary information: The findings of the research

are the sole property of the client. The researcher should not use even a part of it for other
clients' without the prior approval of the exclusive property holder (the client).
Protection against misleading data presentation: Reports should not be presented in

such a way so as to impress greater accuracy than the data warrants. Over-usage of
technical jargon, unnecessary complex analytical procedures, incomplete reporting, etc.
should be avoided. The client has the right to know everything about research design, data
collection, tabulation, analysis and interpretation techniques.
Ethical issues in the treatment of research firms
1. Additional requests: Clients should not make additional requests beyond the purview
of the contract, stating they are paying a large amount on the study. It is not obligatory
of the firm to do so. This leads to conflicts.
2. Financial: Budgets for the study should be indicated so that efforts are not put in to
make a plan that is not actionable.
1 Disclosure of information: If the proprietary data promised by the client is not

disclosed, the completion of the research becomes a difficult task. For example, data
on costs, sales figures, etc.
Corporate espionage: With the competitive environment increasing and the very existence
of the companies under threat, they are resorting to unethical means of information gathering
Le. corporate espionage. All the underlined methods are illegal.
1. Trespassing on the competitor's property.
2. Bribing the competitor's suppliers or employees.
3. Implanting an agent on competitor's pay rolls.
4. Theft of drawings, samples, tenders, documents, etc.
5. Blackmail and the extortion of key employees.
305
Marketing Research
6. Eavesdropping on competitors (for example, telephone tapping).
7. Website hacking.
8. Implanting rumours in the market.
Marketing research is evolving into a full-fledged profession. It is imperative that all parties
concerned in the research should follow certain ethics for the health of this profession.
Sample questionnaire on Customer Satisfaction Survey of credit cards.
XYZ Bank Limited

Customer Satisfaction Survey
About the Questionnaire
• The questionnaire is divided into sections.
• Some sections and some questions may not be applicable to you, in which case, you i
can skip these sections / questions.
4
• Most of the questions require you to merely tick the relevant responses.
• Kindly complete the questionnaire, insert it into the reply-paid envelope enclosed.
and mail it back to us.
SECTION 1 - INTRODUCTION
la. On an overall basis, how satisfied are you with your xyz card ? (Please tick one box)
Very Satisfied Some Neither Somewhat Dissatisfied Very

Satisfied what Satisfied Dissatisfied Dissatisfied
Satisfied Nor
Dissatisfied
A !( . , , ,.
.V f
it.1
IWfTII
306
lb. What is the single most important reason for responding the way you did in Question
I
j
la?
Ic. In the last 12 months, has your xyz card, in terms of its utility and benefits,
Improved substantially Improved a little Remained the same
Become a little worse Become much worse
SECTION 2 - EXPECTATIONS
la. What do you like the most about your xyz card?
lb. What do you dislike about your xyz card ?
If you have been a XYZ Card Owner, prior to January 2000. Please proceed to
Section4A.
The following question relates to your expectations from your xyz card and the service
from the bank, PRIOR TO RECEIVING YOUR CARD.
I What was the single most important reason for going in for an xyz card?
307
Marketing Research
3a. To what extent has your SCB card lived up to your expectations ? (Please tick one
box)
Much better Better About Alitte Much worse

than than the worse than
expected expected same than expected
expected
3b. What is the single most important reason for responding the way you did in
Question 3a?
SECTION 3 - CARD ACQUISITION
If you have been a XYZ Card Owner, prior to January 2000. Please proceed to
Section 4 A
1 a. Taking everything into account, how satisfied are you with the service provided
by the Bank, when applying for your card? (Please tick one box)
• Very satisfied 3b.

• Satisfied
• Somewhat satisfied
• Neither satisfied nor dissatisfied
• Somewhat dissatisfied
• Dissatisfied
308
• Very dissatisfied
1 b. What is the single most important reason for responding the way you did in Question
la?
2a. How long did it take to receive your card from the time you sent in your application?
(please mention the number of days it took to receive your card) _ days.
2b. Ideally, how much time should it have taken to receive your card? ( please mention
the ideal number of days it should have taken to receive your card) _
days.
3a. How satisfied are you with the time taken to receive your card from the time you
submitted your application (please tick one)
• Very satisfied
• Satisfied
• Somewhat satisfied •
• Dissatisfied
What is the single most important reason for responding the way you did in
Question 3a?
309
Marketing Research
SECTION 4 A - CREDIT LIMITS
1 a. Overall, how satisfied are you with the credit limit provided against your XYZ cards
(please tick one box)
• Very satisfied
• Satisfied
»
• Dissatisfied
la?
SECTION 4 B - CARD USAGE
la. On an overall basis, how satisfied are you with the acceptability of your SCB
card ? (That is, in terms of the number and types of establishments that accept your
card. Please tick one box)
• Very satisfied
• Satisfied <
f
• Somewhat satisfied s*
• Dissatisfied
310 • Very dissatisfied
lb. What is the single most imrjortant reason for responding the way you did in Question
la?
2a. Was your card nolaccepted at any establishment ? (Please tick one box)
Yes No (Please proceed to Questions)
2b. Which type(s) of establishments refused to accept your XYZ card ? (Please tick the
relevant boxes, and mention details-viz., name and city)
Garment/Apparel /Textile store Details _____________________________

Consumer Durable outlet ________________________________________
[
H o t e l _ _ _ ________________________________________________
Petrol Pump ... • __________________________________________
Airline _____________________________________________________
Restaurant
Departmental Store.
Others (specify) __
3. Could you suggest any other outlets at / purposes for which you would like to
use j your XYZ card?
r i. _____________
2. __________________
3. ______________________________
4. ____________________ ; ________
5. _______________________________
6. _______________________________
311
Marketing Research
4. Please tell us the degree to which each of the following best describes your
experience, at a merchant outlet, while transacting with your XYZ card-(Please tick
one box for each statement)
Always Frequently Some-times Rarely Neve Does not

r apply
I was able to
complete my
transaction
The transaction was
completed in a
reasonable time
I paid the transaction
amount with no
surcharge
I could transact even
for a very low amount
5. For what purposes do you primarily use your XYZ card ? (Please indicate a
percentage break-up as indicated, in both examples. Please note, your own
answer may be different.)
312
Example 1 Your Answer
Business / Office-related 40% Business / Office-related
Personal 60% Personal
Total 100 Total 100
In India 40 % In India
Abroad 60% Abroad
Total 100 Total 100
Airlines 20% Airlines
Hotels 15% Hotels
Restaurants 20% Restaurants
Buying Durables 05 Buying Durables _

(e.g. washing machines etc.) (e.g. washing machines etc.)
Retail Outlets 05 Retail Outlets -
(e.g. apparel etc.) (e.g. apparel etc.)
Jewellery 05 Jewellery w
Drawing Cash 05 Drawing Cash __

Pertol Pumps Pertrol Pumps _
Others 25 Others __
Total 100 Total 100
6. Taking all your cards into account, please indicate which of these figures best describes
your spending pattern per month, on an average, with your card(s)? (Please tick one
box)
Less than Rs. 1,000 Rs. 1,000-Rs. 2,000 Rs. 2,001 - Rs. 3,000
Rs. 3,001 - Rs. 4,000 Rs. 4,001 - Rs. 5,000 Rs. 5,001 - Rs. 7,500
Rs. 7,501 - Rs. 10,000 Rs.lO,001-Rs.l5,000 Rs. 15,000 +
If you do not own another Card, please proceed to Section 4c.
7. Of all the cards that you own, please indicate, on an average, what percentage of
your monthly card-related expenditure is accounted for by your xyz card?
And, on an average, what percentage of your monthly card-related expenditure do

each of your other cards account for? (Please indicate a percentage break-up)
100 Card % of total card expenditure

1. XYZ card
2. ________
100 3. ________
4. ________
5. ________
31
3
Marketing Research
SECTION 4 C - FLEXIBLE PAYMENT FACILITY
1. Have you ever had a problem with the flexible payment option/revolving
credit facility ? (Please tick one box)
Yes No (PLEASE PROCEED TO SECTION 4D)
If yes, please indicate the nature of the most recent problem in the space provided
below.
SECTION 4 D - STATEMENTS
la. On an overall basis, how satisfied are you with the statements of accounts that you
receive from the bank? (Please tick one box)
• Very satisfied
• Satisfied
• Dissatisfied
Ib. What is the single most important reason for responding the way you did in Question ^
la?
314
Based on your experience, please indicate whether you agree or disagree with each
of the following : (Please tick one box for each statement as applicable)
Always Frequently Some-times Rarely Never Does not apply
I receive my
statements in
time
The statement
is accurate
The statement
is easy to
understand
The statement -
is in a presen-
table form
I get enough
time to make
my payments
before the due
date
SECTION 4 E - PAYMENTS
I. How do you make payments towards the charges against your XYZ card
? (Please tick the appropriate boxes)
Use cheque collection box(es)
Pay / Have it paid at the Bank

Bymail
Others (Specify) _________________________________________________
2a. On an overall basis, how satisfied are you with the service provided by the bank
when you make your payment (Please tick one box)
Very Satisfied Some Neither Somewhat Dissatisfied

Very
Satisfied what Satisfied Dissatisfied Dissatisfied
Satisfied Nor
Dissatisfied
315
Marketing Research
2b. What is the single most important reason for responding the way you did in Question
2a?
3. Please indicate your experience with each of the following statements, keeping
in mind the last one year. (Please tick one box for each statement)
Always Frequently Some-times Rarely Never Does not

apply
My payments are
credited in time
I am able to find
collection boxes at
convenient locations
My payments made
for two /dual cards
are reflected correctly
in my statements
4a. In the last one year, have you faced any problems with respect to making payments
towards the charges incurred against your card? (Please tick one box)
Yes No (PLEASE PROCEED TO SECTION 5)
4b. When did you last have a problem ?
In last 3 months Between 4-6 months ago More than 6 months
ago
4c. Please describe the most recent problem briefly.
316
4d. Was the problem solved to your satisfaction?
Yes No
SECTION 5 - ENQUIRY
la. In the last 6 months, did you ever contact the bank for any purpose, other than for
remitting your payments and enquiring about your account? (Please tick one box)
No Yes (Please Proceed to the next

Sections)
Ib. How did you contact the bank? (Please tick the relevant
boxes) Telephoned the bank Wrote to the bank ByEmail Visited the
bank
Ic. Briefly indicate the nature of enquiry
f
1 d. Kindly indicate whether any of these statements reflected your experience the last
time you contacted the bank. (Please tick one box for each statement)
Agree Disagree Does not apply
Telephone
I was able to get through easily The
phone was picked up promptly I was
not put on hold for a long time
I was put on to the right person within

a reasonable time
317
Marketing Research
The person I spoke to was polite The

person I spoke to was friendly I was
able to get the information that I
wanted / have my query attended to
Mail
I
I received a prompt reply The
letter was politely worded
I was able to get the information that I
In Person
It took a long time before I met

the right person
I was made to wait for a long time
The person I spoke to was polite The
person I spoke to was friendly I was
able to get the information that I
My query was resolved promptly
without follow ups
2a. On an overall basis, how satisfied are you with the service provided by the
bank when you last made an enquiry (Please tick one box)
Very satisifed
Satisfied
Somewhat satisfied
Neither Satisfied nor Dissatisifed
Somewhat satisfied
Dissatisfied
318 Very Dissatisfied

2b. What is the single most important reason for responding the way you did in Question
2a?
SECTION 6 - CASH WITHDRAWAL & TELEDRAFT FACILITY
1. Have you ever used the cash withdrawal facility with your XYZ card ? (Please tick
one box)
Yes No (Please proceed to Question

4)
2a. On an overall basis, how satisfied are you with the service provided by the bank,
when you last made a cash withdrawal? (Please tick one box)
1
Very Satisfied
Satisfied
Somewhat satisfied Neither
Satisfied nor Dissatisfied
Somewhat satisfied Dissatisfied
Very Dissatisfied 1*2b. What is the single most important reason for responding the way
you did in Question
2a?
3a. How long did it take to complete your transaction? (Please mention the time tf
taken to complete the transaction, under the relevant head, namely, Branch or j
ATM)
ATM Seconds Branch/Manual Seconds
3b. Ideally, how much time should it have taken? (Please mention the ideal time it
should have taken to complete the transaction, under the relevant head, namely,
branch or ATM)
319
Marketing Research
ATM Seconds Branch/Manual Seconds
4. Have you ever used the Teledraft facility? (Please tick
one) Yes No
SECTION 7 - SATISFACTION WITH OVERALL SERVICE
1 a. Taking everything into account, how satisfied are you with the overall quality
of service provided by XYZ? (Please tick one box)
Very satisfied
Satisfied
Somewhat satisfied
Neither Satisfied nor Dissatisfied
Somewhat satisfied
Dissatisfied
Very Dissatisfied
la?
1 c. In the last 12 months, has the overall quality of service provided by XYZ
Improved substantially
Improved a little
Remained the same
Become a little worse
Become much worse
320
SECTION 8 - PROBLEMS WITH XYZ CARD
I la. On an overall basis, how satisfied are you with the service provided by the
bank
I when you last had a problem? (Please tick one box)
Very satisfied
I Satisfied
r
P?
Somewhat satisfied
| Neither Satisfied nor Dissatisfied
jn!
Somewhat satisfied
Dissatisfied
Very Dissatisfied «>Ib. What is the single most important reason for responding the
way you did in Question
2a?
2a. In the last 1 year, did you have any problem with your xyz card ?
Yes No (Please proceed to the next section )
2b. When did you have a problem with respect to your XYZ card ?
In last 3 months 4-6 months ago 6- 12 months ago > 12 months ago
D
2c. How many times have you had a problem with the bank ? (Please tick one box)
Once Twice Thrice More than three times
2d. What was the most recent problem that you had with xyz ?
Marketing Research •
"V
2e. In which of these categories, would you place your most recent problem?
(Please tick the relevant boxes)
Statement related Payment related Cash Withdrawal related

Transaction related Enquiry related staff related
Others (please specify)
2f. How would you regard this problem? (Please tick one box)
An irritant A minor problem A major problem
2g. Did you contact the bank about the problem? (Please tick one box)
Yes No (Please proceed to next section)
2h. How did you contact the bank? (Please tick one box)
Visited the bank Wrote to the bank
Telephoned the bank Bye-mail
2i. Did you get a response from the bank? (Please tick one box) I
got an immediate response from the bank I got a response from the
bank after one or more reminders I have still not got a response
from the bank
2j. Has this problem been resolved? (Please tick one box)
Yes, it has been resolved The bank is still trying to solve the
problem The bank was unable to solve the problem to my
satisfaction
2k. If the problem has been resolved, how long did it take to resolve the problem? days
Ideally, how long should it have taken? ___ days
3a. Have any of your problems been of a repetitive nature in the last one year
(Please tick one box)
Yes No (Please proceed to the next section)
322
Marketing Research
SECTION 10 - SUGGESTIONS
1 a. Please give ONE suggestion that will add value to your SCB card.
1 b. Please give us ONE suggestion that will help us improve the quality of service delivered
byXYZ.
SECTION 11 - ABOUT YOURSELF - PROFILE & CLASSIFICATION DATA
This section seeks to obtain demographic and classification data for STATISTICAL
PURPOSES ONLY. Information provided will be treated as confidential.
la. Age in years (Please tick one box)

Less than 25 25-35
46-55 55 +
b. Annual Income (Personal & Household) (Please tick one box under each column)
Income in Rs./Yr. Personal Household
Less than Rs. 60,000
Between Rs. 60,000 - 75,000
Between Rs. 75,000 - Rs. 1 Lakh ;
324
Between Rs. 1 Lakh - Rs. 1 .5 Lakh

Between Rs. 1 .5 Lakh - Rs. 2 Lakhs
Between Rs. 2 Lakhs - Rs. 2.5 Lakhs
Between Rs. 2.5 Lakhs - Rs. 3 Lakhs
Between Rs. 3 Lakhs - Rs. 5 Lakhs Rs.
5 Lakhs +
c Education (Please tick one box)
Graduate Post Graduate Doctorate Others
d. Occupation (Please tick one box)
Business Salaried-Pvt. Sector Salaried-Public Sector/GovtOrgn.
Self-Employed Professional (e.g. Doctor, Architect, Lawyer, Engg. Consultant,
Management Consultant etc.) O t h e r s ( s p e c i f y ) _ _ _ _ _
_ _
e. N u m b e r o f E a r n i n g M e m b e r s i n t h e h o u s e h o l d : _
f. Durables owned and intend to own in the next 1 year. (Please tick the
relevant boxes under each column)
Own currently Intend to acquire In next one year

Colour T.V.
DVD/CD player
Frost Free Fridge
Washing Machine
Vacuum Cleaner
Microwave oven
Dish washer
Air Conditioner
Cellular phone
HI-FI music system
325
Marketing Research
Own currently Intend to acquire In next one year

Laser Disc Cam i
corders/Handycam Digital
camera Personal
Computer Laptops
g.. Vehicle Ownership (Please tick the relevant boxes)

None 2 Wheeler 4 Wheeler
Number of 4 Wheelers Owned (Please tick one box)
1 2+ None
2. Travel abroad (Please tick one box for each row)

Yes No
j
Everbeen
In last 1 year
Likely to travel in next 1 year
,*
Thank you for taking the time to complete this questionnaire
May we include your name with your responses to the data in this survey ?
(If you wish to keep your responses anonymous, we will ensure that your wishes are
respected and adhered to.)
Yes No
Please indicate your complete name & address for updating of our records
Name (Mr/Ms)
Address . ^
City Pin
Card Number
326
15.7 SUMMARY
In this unit, we have learnt about the modern approach of survey methods through the
internet era. We have also seen how services play an important role in the service industry
in the questionnaire designed for the bank's credit cards. One can understand from the
different sections how systematically all the information is gathered and analysed so as to
improve on the aspects which they feel more attention is to given on the particular area in
which they are lacking. But still, as these are not a foolproof approach, there is very much
less authenticity in the internet survey still fifty percent is taken on average as correct and
genuine answers given by the respondents.
15.8 KEY WORDS ________________________________________

;
Availed: be of use
Ambience: surroundings or atmosphere
Confectionary: sweets
Abuse: misuse
Anonymity: of unknown name, or authorship
Disclosure: expose
15.9 SELF-ASSESSMENT QUESTIONS _____________________________

Q1. In this era of the internet, do you feel the on line survey is beneficial for the
survey? Comment your answer.
Q2. What is in your opinion, any five disadvantages of an online survey?
Q3. An airline industry wants to do a survey on customer satisfaction. They need your
help for designing the questionnaire. How will you do it?
327

Marketing Research Tool

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Marketing Research Tool

Uploaded by

Copyright:

Available Formats

CONTENTS

1. Fundamentals of Market Research

Application and Limitations 19

4. Questionnaire - Design Development 61

Data Collection Primary / Secondary 117

Data Preparation and Processing 139

8. Data Analysis and Interpretation 159

10. Research Process and Design 193

11. Specialised Techniques in Market Research 209

12. Market and Sales Analysis Research 231

14. Presentation of Research Findings 279

1.1 THE MARKETING ENVIRONMENT

1.2 MARKET RESEARCH

Another simpler definition can be put in this way.

Research can be classified into two broad categories:

i) Basic research ii) Applied research

Applied research can be divided into two categories:

Problem-oriented research, on the other hand, is concerned with a class of issues or

1.3 THE NEED FOR MARKET RESEARCH

There are several factors which have contributed to the research:

Multinational company Whirlpool Asia lays considerable emphasis on Market research.

Market research is a systematic and objective study of problems pertaining to the

B. The Growth of Market research

Third, the emergence of specialists such as statisticians, psychologists and behavioural

C. Market research for Industrial Goods

Geographic concentration implies that there is a general tendency among many

4. Purchasing procedures and organisation '

The relationship between the manufacturer and buyer of industrial goods is

1.5 THREATS TO MARKET RESEARCH

1.6 MARKET INFORMATION SYSTEM (MIS)

Recurrent information: This type of information is provided on periodic basis, which

These are or may be supplied on a weekly or monthly basis.

Monitoring information: This type of information is obtained from regular scanning of

Requested information: Requested information as the name suggests, this is sought by

1.7 MARKET DECISION SUPPORT SYSTEM (MDSS) __________

MDS S can be classified into five components

Modelling is the last component of MDSS. Models are mathematical formulations.

"Market research is undertaken to guide managers in the analysis of marketing problems."

1.8 SUMMARY _________________________________________ _ _ _ _

1.9 KEYWORDS _____________________ _ _ _ ^ _ _____________

Entrepreneur: a person who undertakes a commercial venture

a) Will research, play an important role in the development of Sony India?

Research management concentrates mainly on the processes, projects, personnel, finances

A. Importance of research management

B. Qualities of a market research manager

• innovative and always be prepared to look at any problem from a fresh

• familiar with recent developments in particular industries, especially the problems

One of the most crucial decisions to be taken by management is whether market

C. Organising market research function

consultancy firm, as the requirements and emphasis on research will be different. A

Coming to the organisation of research activity, several alternatives are open to a

a. to set up a formalised market research department consisting of some full-time

c. to assign the responsibility of undertaking market research to an outside individual

If a company has decided to set up a separate market research department, it is very

a) List any five qualities of a good research manager.

b) Why is research regarded as useful to marketing management?

D. Selection of market research

E. Selection of research agency

Criteria for hiring outside agencies

• Timely submission of reports

• Experience of the supplier

• Reputation of the agency

• Test marketing studies