BBRS4103

MARKETING
RESEARCH
Assoc Prof Dr Jamil Bojei
Dr Che Aniza Che Wel
Dr Oh Teik Hai
Project Directors: Prof Dr Mansor Fadzil
Prof Dr Zakaria Ismail
Open University Malaysia

Module Writers: Assoc Prof Dr Jamil Bojei

Universiti Putra Malaysia
Dr Che Aniza Che Wel
Universiti Kebangsaan Malaysia
Dr Oh Teik Hai

Moderator: Assoc Prof Dr Aliah Hanim Md Salleh

Universiti Kebangsaan Malaysia
Dr Wardah Mohamad
Open University Malaysia

Developed by: Centre for Instructional Design and Technology

Open University Malaysia

Printed by: Meteor Doc. Sdn. Bhd.

Lot 47-48, Jalan SR 1/9, Seksyen 9,
Jalan Serdang Raya, Taman Serdang Raya,
43300 Seri Kembangan, Selangor Darul Ehsan

First Edition, January 2005

Second Edition, December 2012 (rs)

Copyright © Open University Malaysia (OUM), December 2012, BBRS4103

All rights reserved. No part of this work may be reproduced in any form or by any means without
the written permission of the President, Open University Malaysia (OUM).
Table of Contents
Course Guide xiăxvi

Topic 1 Introduction to Marketing Research 1

1.1 Definitions of Marketing Research 2
1.2 Classification of Marketing Research 3
1.3 The Marketing Research Process 5
1.4 An Overview of the Marketing Research Industry 6
1.4.1 Research Agencies In Malaysia 6
1.5 Selecting the Research Supplier 9
1.6 Careers in Marketing Research 10
1.7 Economic Systems 11
1.7.1 The Internal Report System 12
1.7.2 Marketing Information System (MIS) 12
1.7.3 Marketing Decision Support System (DSS) 13
1.7.4 Marketing Research Systems 13
1.8 Ethics and Marketing Research 14
Summary 16
Key Terms 17

Topic 2 Defining the Marketing Research Problem 18

2.1 Defining the Problem 19
2.2 The Process of Defining the Problem and Developing
an Approach 20
2.2.1 Discussion with the Decision Maker 20
2.2.2 Interview with Industry Experts 21
2.2.3 Secondary Data Analysis 21
2.2.4 Translating the Marketing Decision Problem into
a Marketing Research Problem 21
2.3 The Nature of the Problem 22
2.4 Marketing Decision Problems and Marketing Research
Problems 23
2.5 Defining the Marketing Research Problem 23
2.6 Components of Approach to Research 24
Summary 29
Key Terms 30
iv  TABLE OF CONTENTS

Topic 3 Research Design 31

3.1 What is Research Design? 32
3.2 Basic Research Designs 33
3.2.1 Exploratory Research 33
3.2.2 Conclusive Research 35
3.3 Relationships among Exploratory, Descriptive and
Causal Research 40
3.4 Tasks Involved in Formulating a Research Design 41
3.5 Potential Sources of Error 41
3.5.1 Sampling Error 42
3.5.2 Non-sampling Error 43
3.6 Budgeting and Scheduling the Project 45
3.7 Marketing Research Proposal 45
Summary 48
Key Terms 48

Topic 4 Data Collection Method and Sample Design 49

4.1 Secondary Data 50
4.1.1 Primary versus Secondary Data 50
4.1.2 Advantages and Disadvantages of Secondary
Data 51
4.1.3 Criteria for Evaluating Secondary Data 53
4.1.4 Internal Secondary Data and Database Marketing 54
4.1.5 Published Sources of External Secondary Data 56
4.2 Qualitative Research 57
4.2.1 Primary Data: Qualitative versus Quantitative
Research 57
4.2.2 Classification of Qualitative Research Procedures 58
4.3 Quantitative Research 64
4.3.1 Survey Methods Classified by Mode of
Administration 64
4.3.2 A Person-administered Survey 65
4.3.3 A Computer-administered Survey 65
4.3.4 A Self-administered Survey 66
4.3.5 Criteria for Selecting Survey Methods 67
4.3.6 Improving Survey Response Rates 70
4.4 Observation Methods 71
4.4.1 Relative Advantages and Disadvantages of
Observation Methods versus Survey Methods 72
4.4.2 Comparison of Survey and Observation Methods 73
4.5 Concept of Causality 75
4.5.1 Conditions for Causality 75
 TABLE OF CONTENTS  v

4.6 What is Experimentation? 76

4.6.1 Validity of Experimentation 77
4.6.2 A Classification of Experimental Designs 77
4.6.3 Laboratory versus Field Experiments 78
4.6.4 Limitations of Experimentation 79
Summary 80
Key Terms 82

Topic 5 Measurement and Scaling 83

5.1 Measurement and Scaling 84
5.1.1 Primary Scales and Level of Measurement 85
5.2 Comparative Scaling Techniques 89
5.2.1 Paired Comparison Scaling 89
5.2.2 Rank Order Scaling 89
5.2.3 Constant Sum Scaling 90
5.3 Non-comparative Scaling Techniques 91
5.3.1 Continuous Scaling 91
5.3.2 Itemised Rating Scales 92
5.3.3 Non-comparative Itemised Rating Scale Decision 93
5.3.4 Multi-items Scales 95
5.3.5 Scale Evaluation 95
Case Study 97
Summary 99
Key Terms 100

Topic 6 Questionnaire Design 101

6.1 The Importance of Questionnaires 102
6.2 Questionnaire Design Process 103
6.2.1 Specify the Information Needed 103
6.2.2 Type of Interviewing Method 104
6.2.3 Individual Question Content 104
6.2.4 Overcoming Inability to Answer 104
6.2.5 Overcoming Unwillingness to Answer 105
6.2.6 Choosing the Question Structure 105
6.2.7 Choosing the Question Wording 105
6.2.8 Order of the Questions 107
6.2.9 Form and Layout 107
6.2.10 Reproduction of the Questionnaires 107
6.2.11 Pre-testing 108
6.3 Observational Forms 108
Case Study 111
Summary 113
Key Terms 113
vi  TABLE OF CONTENTS

Topic 7 Sampling 114

7.1 Sampling Design and Procedures 115
7.2 Sampling Design Process 116
7.2.1 Define the Target Population 117
7.2.2 Determine a Sampling Frame 117
7.2.3 Select a Sampling Technique 117
7.2.4 Determine the Sample Size 118
7.2.5 Execute the Sampling Process 119
7.3 Classification of Sampling Techniques 119
7.3.1 Non-probability Sampling Techniques 120
7.3.2 Probability Sampling Techniques 121
7.3.3 Choosing Non-probability versus Probability
Sampling 122
7.4 Initial and Final Sample Size Determination 122
7.4.1 Definition of Symbols 123
7.4.2 Sampling Distribution 124
7.4.3 Statistical Approach to Determine Sample Size 125
7.4.4 Confidence Interval Approach 125
7.4.5 Adjusting the Statistically Determined Sample Size 128
7.4.6 Special Situations in Sample Size Determination 129
Case Study 129
Summary 131
Key Terms 132

Topic 8 Field Work 133

8.1 Nature of Field Work 134
8.2 Field Work/Data Collection Process 134
8.3 Selection of Field Workers 135
8.4 Training of Field Workers 136
8.4.1 Making the Initial Contact 136
8.4.2 Asking Questions 136
8.4.3 Probing 137
8.4.4 Recording Answers 137
8.4.5 Terminating the Interview 138
8.5 Supervision of Field Workers 138
8.5.1 Quality Control and Editing 139
8.5.2 Sampling Control 139
8.5.3 Control of Interviewer Cheating 139
8.5.4 Central Office Control 140
8.6 Validation of Field Workers 140
8.7 Evaluation of Field Workers 140
Summary 142
Key Terms 142
 TABLE OF CONTENTS  vii

Topic 9 Data Preparation and Analysis Strategy 143

9.1 Data Preparation Process 144
9.2 Questionnaire Checking 145
9.3 Editing 145
9.4 Coding 147
9.4.1 Coding Questions 148
9.5 Transcribing 150
9.6 Data Cleaning 151
9.6.1 Consistency Checks 151
9.6.2 Treatment of Missing Responses 152
9.7 Selecting a Data Analysis Strategy 153
9.8 Strategies for Qualitative Data 155
Summary 159
Key Terms 160

Topic 10 Report Preparation and Presentation 161

10.1 Importance of the Report and Presentation 162
10.2 Questionnaire Checking 162
10.3 Report Preparation 163
10.3.1 Report Format 164
10.3.2 Report Writing 168
10.3.3 Guidelines for Tables 169
10.3.4 Guidelines for Graphs 170
10.3.5 Oral Presentation and Dissemination 171
10.3.6 Research Follow-up 172
Summary 176
Key Terms 176
viii  TABLE OF CONTENTS
COURSE GUIDE
COURSE GUIDE DESCRIPTION
You must read this Course Guide carefully from the beginning to the end. It tells
you briefly what the course is about and how you can work your way through
the course material. It also suggests the amount of time you are likely to spend in
order to complete the course successfully. Please continue referring to the Course
Guide as you go through the course material because it will help you to clarify
important study components or points that you might miss or overlook.

INTRODUCTION
BBRS4103 Marketing Research is one of the courses offered by the Faculty of
Business and Management at Open University Malaysia (OUM). This course is
worth 3-credit hours and should be covered over 8 to 15 weeks.

COURSE AUDIENCE
This is an elective course for learners undergoing the Bachelor of Business
Administration programme, specialising in Marketing.

As an open and distance learner, you should be acquainted with learning

independently and be able to optimise the learning modes and environment
available to you. Before you begin this course, please ensure that you are aware
of the course material, the course requirements and how the course is conducted.

STUDY SCHEDULE
It is a standard OUM practice that learners accumulate 40 study hours for every
credit hour. As such, for a three-credit-hour course, you are expected to spend
120 study hours. Table 1 gives an estimation of how the 120 study hours could be
accumulated.
xii  COURSE GUIDE

Table 1: Estimation of Time Accumulation of Study Hours

Study
Study Activities
Hours
Briefly go through the course content and participate in initial 3
discussion
Study the module 60
Attend 3 to 5 tutorial sessions 10
Online participation 12
Revision 15
Assignment(s), Test(s) and Examination(s) 20
TOTAL STUDY HOURS 120

LEARNING OUTCOMES
By the end of this course, you should be able to:

1. Discuss the process of marketing research and describe the development

approach to marketing research problems;

2. Explain the exploratory research design and discuss the pros and cons of
secondary data and qualitative research;

3. Differentiate between descriptive and causal design in marketing research;

4. Justify the various scale formats for measurement and how to construct them;
and

5. Design a marketing research plan and conduct the research.

COURSE SYNOPSIS
This course is divided into ten topics. The synopsis for each topic is presented
below:

Topic 1 will distinguish the two types of research (problem-identification

research and problem-solving research) and describe the framework for
conducting the marketing research as well as the steps involved in the marketing
research process. This topic also explains the nature of marketing research and its
 COURSE GUIDE  xiii

role in designing and implementing successful marketing programmes in the

marketplace. Learners will also be exposed to the types and roles of research
suppliers and research agencies in Malaysia. The discussion will also include the
role of marketing research to support the marketing information systems (MIS)
and decision support systems (DSS).

Topic 2 covers the first two steps in the marketing research process: defining
marketing research problems and developing an approach to the problems.
There will be an explanation on the importance of and the process for
determining marketing research problems, besides the tasks involved in the
problem definition process. This topic also discusses the environmental factors
affecting problem definition and the distinction between management decision
problems and marketing research problems.

Topic 3 discusses the types of research designs that are available to the marketing
researcher. Learners will be exposed to the appropriateness of the three basic
research designs: exploratory, descriptive and causal research. This topic also
describes the major errors in a research design, the elements of a marketing
research project and the role of information technology in improving the
components of research design and minimising research error.

Topic 4 explains two aspects of exploratory research design: secondary data and
qualitative research. It discusses the nature of secondary data and distinguishes
secondary data from primary data. Various sources of secondary data are also
explained in this topic. In addition, learners will be exposed to the classification
of qualitative research procedures. Also, descriptive research design, which is
undertaken to describe answers to questions of who, what, where, when, and
how, is discussed. This topic will discuss and classify survey methods and
describe the various interviewing methods. Learners will be exposed to the
strengths and limitations of each survey method, methods of increasing the
survey response rate, and the advantages and disadvantages of observation and
survey methods. An overview of causal research design and experimentation is
analysed. It explains the concept of causality in marketing research, differentiates
the types of validity and describes the classification of experimental design. The
learners will also be exposed to laboratory and field experiments.

Topic 5 describes the concepts of scaling and measurement. It will distinguish the
four primary scales of measurements: nominal, ordinal, interval and ratio. It will
also compare and examine each type of comparative and non-comparative
scaling technique.
xiv  COURSE GUIDE

Topic 6 focuses on designing the forms upon which data will be recorded, which
includes questionnaires and observation forms. This topic will explain the
purpose of questionnaires and their objectives and describe the process of
designing questionnaires.

Topic 7 will explain how sampling is used in marketing research. This topic
differentiates a sample from a census, discusses the sampling process and defines
the target population for a specific study. Learners will be exposed to methods
of determining the sampling frame, selecting the sampling techniques and
determining the appropriate sample size. This topic will also classify the different
sampling techniques.

Topic 8 will focus on the fourth step in the marketing research process, which is
field work. It describes the field work process and explains the selection, training,
supervision, validation and evaluation of field workers.

Topic 9 discusses the nature and scope of the data preparation process, describes
questionnaire checking and editing activities as well as explains and provides
guidelines for coding questionnaires. It will also discuss the data cleaning
process and describe the procedures for selecting a data analysis strategy.

Topic 10 will explain the final step of the marketing research process, which is
report preparation and presentation. It will discuss the requirements of report
preparation, describe the research report and presentation process and explain
the importance of research follow-up.

TEXT ARRANGEMENT GUIDE

Before you go through this module, it is important that you note the text
arrangement. Understanding the text arrangement will help you to organise your
study of this course in a more objective and effective way. Generally, the text
arrangement for each topic is as follows:

Learning Outcomes: This section refers to what you should achieve after you
have completely covered a topic. As you go through each topic, you should
frequently refer to these learning outcomes. By doing this, you can continuously
gauge your understanding of the topic.

Self-Check: This component of the module is inserted at strategic locations

throughout the module. It may be inserted after one sub-section or a few sub-
sections. It usually comes in the form of a question. When you come across this
component, try to reflect on what you have already learnt thus far. By attempting
 COURSE GUIDE  xv

to answer the question, you should be able to gauge how well you have
understood the sub-section(s). Most of the time, the answers to the questions can
be found directly from the module itself.

Activity: Like Self-Check, the Activity component is also placed at various

locations or junctures throughout the module. This component may require you to
solve questions, explore short case studies, or conduct an observation or research.
It may even require you to evaluate a given scenario. When you come across an
Activity, you should try to reflect on what you have gathered from the module and
apply it to real situations. You should, at the same time, engage in higher order
thinking where you might be required to analyse, synthesise and evaluate instead
of only having to recall and define.

Summary: You will find this component at the end of each topic. This component
helps you to recap the whole topic. By going through the summary, you should
be able to gauge your knowledge retention level. Should you find points in the
summary that you do not fully understand, it would be a good idea for you to
revisit the details in the module.

Key Terms: This component can be found at the end of each topic. You should go
through this component to remind yourself of important terms or jargon used
throughout the module. Should you find terms here that you are not able to
explain, you should look for them in the module.

References: The References section is where a list of relevant and useful

textbooks, journals, articles, electronic contents or sources can be found. The list
can appear in a few locations such as in the Course Guide (at the References
section), at the end of every topic or at the back of the module. You are
encouraged to read or refer to the suggested sources to obtain the additional
information needed and to enhance your overall understanding of the course.

PRIOR KNOWLEDGE
No prior knowledge is required.

ASSESSMENT METHOD
Please refer to myVLE.
xvi  COURSE GUIDE

REFERENCES
Aaker, D. A., Kumar, V., & Day, G. S. (2009). Marketing research (10th ed.).
New York, USA: John Wiley & Sons.

Burns, A. C., & Bush, E. F.. (2006). Marketing research (5th ed). Upper Saddle,
New Jersey, USA: Prentice-Hall, Inc.

Churchill, Jr., G. A. (2001). Basic marketing research, (3rd ed.). Fort Worth, Texas.
USA: The Dryden Press (Harcourt Brace College Publishers).

Churchill, Jr. G. A. (2005). Marketing research: Methodology foundations

(9th ed.). Fort Worth, Texas, USA: The Dryden Press (Harcourt Brace College
Publishers).

Malhotra, N. (2002). Basic marketing research: Applications to contemporary

issues. Upper Saddle River, New Jersey, USA: Prentice-Hall, Inc.

Neuman, W. L. (2003). Social research methods (5th ed.). New York, USA:
Pearson Education, Inc.

Saunders, M., Lewis, P., & Thornhill, A. (2009). Research methods for business
students. Edinburgh Gate, England: Pearson Education Limited.

TAN SRI DR ABDULLAH SANUSI (TSDAS)

DIGITAL LIBRARY
The TSDAS Digital Library has a wide range of print and online resources for
the use of its learners. This comprehensive digital library, which is accessible
through the OUM portal, provides access to more than 30 online databases
comprising e-journals, e-theses, e-books and more. Examples of databases
available are EBSCOhost, ProQuest, SpringerLink, Books24x7, InfoSci Books,
Emerald Management Plus and Ebrary Electronic Books. As an OUM learner,
you are encouraged to make full use of the resources available through this
library.
Topic  Introduction
to Marketing
1 Research
LEARNING OUTCOMES
By the end of this topic, you should be able to:
1. Define marketing research;
2. Distinguish between problem identification and problem solving
research;
3. Describe the framework for conducting marketing research as well as
the six steps in marketing research process;
4. Explain the nature of marketing research and its role in designing
and implementing successful marketing programmes;
4. Identify the types and roles of research suppliers and the research
agencies in Malaysia; and
5. Assess the role of marketing research to support marketing
information systems (MIS) and decision support systems (DSS).

 INTRODUCTION
Have you ever wondered how a television station knows which programmes are
the most popular and which ones should be cancelled? Have you wondered how
radio stations know which deejays are the most listened to during the day? Ever
asked yourself how businesses know if they need to change their product
features or perhaps offer something new? Well, a lot of factors go into these
decisions. One major factor is marketing research.
2  TOPIC 1 INTRODUCTION TO MARKETING RESEARCH

Marketing research is one of the most important and fascinating tasks in

marketing. This topic will define two broad forms of research:
(a) Research designed to identify problems; and
(b) Research designed to solve problems.

We will also look at six steps in the marketing research process. Then, we will
look at various firms in the marketing research industries as well as the role of
marketing research to support the marketing information systems (MISs) and
decision support systems (DSSs).

1.1 DEFINITIONS OF MARKETING RESEARCH

What do you think marketing research involves?

We define marketing research as a systematic and objective identification,

collection, analysis and dissemination of information for the purpose of assisting
management in decision making related to the identification and solution of
problems (and opportunities) in marketing.

Our definition tells us that marketing research is the process that results in
reporting information which can be used to solve marketing problems such as
setting the best pricing strategy, choosing the best promotional methods and so
on. Therefore, the focus is on the process that results in information that will be
used to make decisions. Ours is not the only definition of marketing research.
The American Marketing Association (AMA) formed a committee several years
ago to establish a definition of marketing research, and formulated the following
definition:

Marketing research is the function that links the consumer, customer, and
public to the marketer through information ă information used to identify and
define marketing opportunities and problems; generate, refine and evaluate
marketing performance; and improve the understanding of marketing as a
process.

Which one of these definitions is correct? They are both right but our definition is
shorter. The AMAÊs definition is not as short because it elaborates on the function
(we call it „the purpose‰) and uses of marketing research.
 TOPIC 1 INTRODUCTION TO MARKETING RESEARCH  3

Marketing research begins with the identification or definition of the research

problem or opportunity. Problems often lead to opportunities in business, so
in the research perspective, the two words are used interchangeably. As an
example, the fact that more and more women go out to work and less time is
spent at home represents a problem as well as an opportunity for producers of
food products. The problem is that when women go to work, they have no time
to cook, which results in a decrease in demand for food that requires a longer
time to cook. However, the opportunity is that they will demand fast food or
instant food.

The range of data collection methods and the sources to collect data vary in
sophistication and complexity. In addition, the methods used depend on the
specific requirement of the project, including budget and time constraints. The
data is then analysed and the results are formally presented to the client.

ACTIVITY 1.1

In your opinion, why is marketing research important for any

organisation?

1.2 CLASSIFICATION OF MARKETING

RESEARCH
Problem-identification research is research undertaken to help identify
problems that are not necessarily apparent on the surface and yet exist or are
likely to arise in the future.

Problem-identification research may be designed to estimate market potential,

or market share; to study brand or company image, market characteristics,
sales analysis, short-range forecasting, or long-range forecasting; or to uncover
business trends. For instance, John Master Sdn. Bhd. may undertake research to
determine the size of the market for menÊs shirts, the projected growth rate and
the market share of major brands. This kind of research may sometimes reveal a
problem. For example, although John MasterÊs sales of menÊs shirts have been
increasing, they have not been keeping pace with the growth of the industry.
Moreover, John Master has been gradually losing its market share to competitors
like Playboy.
4  TOPIC 1 INTRODUCTION TO MARKETING RESEARCH

Problem-identification research is normally conducted by a company to assess

the environment and diagnose problems. Information pertaining to changes in
the environment can be used as an initial indicator of potential opportunities
or problems. For example, food processing firms such as Danone Marketing
(Malaysia) Sdn. Bhd. and Nestle Products Sdn. Bhd. have discovered
opportunities in consumers' increasing concern about health, weight and diet.

Once the problem or opportunity has been identified, the firm undertakes
problem solving research to address the problem. Problem-solving research is
done to help solve specific marketing problems. This type of research normally
addresses segmentation, product, pricing, promotion and distribution. Going
back to our earlier example of food processing firms, these types of studies
would determine alternative ways of appealing to health-conscious consumers.
These firms would introduce new foods to the market that are low in cholesterol,
fat, sodium and sugar. If research indicates that low-cholesterol biscuits are the
most popular, what special formulation of the biscuit would best serve the
market? What price should be used? How should the biscuits be promoted and
distributed? Figure 1.1 summarises the classification of marketing research.

Figure 1.1: Classification of marketing research

Source: Malhotra, N. (2010). Marketing research, An applied orientation, (6th ed.).
Upper Saddle River, New Jersey, USA: Prentice-Hall, Inc.
 TOPIC 1 INTRODUCTION TO MARKETING RESEARCH  5

1.3 THE MARKETING RESEARCH PROCESS

The marketing research process consists of the following six steps:

(a) Defining the Problem

Defining the problem not only involves trying to understand the purpose
of the study but also trying to understand the background issues that
go along with it. The background issues include such questions as „What
prompted the research?‰ and „What kind of decisions will be made from its
results?‰ It will also address the value of the information that the research
will gather and who will benefit from the research.

(b) Developing an Approach to the Problem

Developing an approach to the problem includes formulating an analytical
framework and models, research questions and hypotheses.

(A hypothesis is an unproven statement or proposition about a factor or

phenomenon that is of interest to the researcher). The development of an
approach will be discussed in greater detail in Topic 2.

(c) Formulating a Research Design

A research design is a framework or blueprint for conducting marketing
research projects. It details the procedures needed to obtain the required
information.

(d) Doing Field Work or Collecting Data

Data collection is accomplished using employees who operate in the field.
Field work can be in the form of surveys or observations. Surveys involve
personal, telephone, mail or electronic interviewing whereas observation
can be either personal or mechanical. Proper selection, training, supervision
and evaluation of the field force are essential to ensure that high quality
data is gathered.

(e) Preparing and Analysing Data

Data preparation involves data-processing steps leading up to an analysis.
This includes editing, coding and transcribing as well as cleaning up the
data. The entire process must then be verified for accuracy.

(f) Preparing and Presenting the Report

The entire project should be documented in a written report that addresses
the specific research questions and describes the approach, the research
design, data collection and data analysis procedures. Finally, the results
and major findings of the research should be presented to the client.
6  TOPIC 1 INTRODUCTION TO MARKETING RESEARCH

SELF-CHECK 1.1

What are the steps in the marketing research process?

1.4 AN OVERVIEW OF THE MARKETING

RESEARCH INDUSTRY
Marketing research has a broad range of applications and plays a crucial
role in the marketing decision-making process. The major goal of marketing
is to identify and satisfy customer needs. To do this, marketing managers
need information about customers, competitors and other forces such as
environmental trends in the marketplace. In order to become competitive, timely
information is needed and marketing research is a tool to obtain the information
required.

In the marketing research industry, the provider of marketing information is

referred to as a research supplier. It has been estimated that firms spend roughly
10 percent of sales on marketing research, whether it is done internally or
externally. An internal supplier is an entity within the firm. Kodak and General
Motors, for example, have research departments of their own. Similarly, AT&T
has an in-house research department that constantly monitors consumer
satisfaction and environmental trends. It also provides research support to AT&T
advertising agencies.

External suppliers are outside firms hired to fulfil the firmÊs marketing research
needs. Large and small firms, commercial and not-for-profit organisations, and
government and educational institutions all purchase research information from
external suppliers. You can get an idea of the type, number of and services
offered by these firms by looking at some online directories of marketing
research firms.

Go to www.greenbook.org and click on the picture of the Greenbook. Explore

this website to get a better understanding of how to classify external supplier
firms. Also visit another directory listing at www.quirks.com.

1.4.1 Research Agencies in Malaysia

(a) ACNielsen (MÊsia) Sdn. Bhd. (http://www.acnielsen.com.my) is a famous
external supplier that offers a wide range of research services to its
consumers. ACNielsen is the worldÊs leading provider of market research,
 TOPIC 1 INTRODUCTION TO MARKETING RESEARCH  7

information and analysis to the customer products and services industries,

as well as to government and social services. More than 9,000 clients in over
100 countries rely on ACNielsenÊs dedicated professionals to measure
competitive marketplace dynamics, to understand customer attitudes and
behaviour and to develop advanced analytical insights that generate
increased sales and profits.

ACNielsen Malaysia provides three principal market research services:

(i) Retail measurement, including continuous tracking of consumer

purchases at the point of sales through scanning technology
and in-store audit. Clients receive detailed information on actual
purchases, market share and advertising effectiveness.

(ii) Customised research, including quantitative and qualitative studies

that generate information and insights into customersÊ attitudes and
purchasing behaviour, customer satisfaction, brand awareness and
advertising effectiveness.

(iii) Information for general use, in which ACNielsen provides ongoing

feedback to the industries that they work with through a variety of
publications produced locally in each country; they also work with
the press to provide information on top line category trends,
consumer and audience behaviour. These items can be found in the
Market Insights section in the ACNielsen web page or in General
News areas updated regularly. Any information on the performance
of specific programmes, brands or suppliers is confidential and not
available to the public, although the clients might release data at their
own discretion.

In general, ACNielsen has been collecting information in the Asia Pacific

region for over 80 years, providing marketers and business professionals
with invaluable information and insight into consumersÊ attitudes and
motivations, purchase habits, brand preferences and media consumption
patterns. Through the integration of ACNielsenÊs core services ă customised
research and retail and media measurement ă the company offers
unparalleled insights into consumer behaviour and attitudes and the
diverse market within the Asia Pacific region.

As a service to their clients and the industry in general, ACNielsen

publishes annual trend and market information on the leading grocery
brands, as well as the grocery, pharmaceutical and media industries.
8  TOPIC 1 INTRODUCTION TO MARKETING RESEARCH

(b) Department of Statistics Malaysia (http://www.statistics.gov.my) is the

government institution that conducts various studies for government
and public use. The main functions of the department, as provided in
the Statistics Act 1965 (Revised ă 1989), are the collection, interpretation
and dissemination of statistics for the purpose of formulating policies for
national development and planning administration.

Under this Act, the Department of Statistics conducts various sample

surveys and censuses. It has also conducted the Population and Housing
Census for the years 1970, 1980, 1990, 2000 and 2010 under the Census Act
1960.

The increasingly important and critical role of the Department of Statistics,

in meeting the challenges posed by the growing demand and expectation of
users, is in line with the tremendous pace of development in managing and
planning of the national economy as well as in the use of information
technology.

Thus, the department has reorganised its programmes and work

procedures to enable it to effectively utilise its resources for the provision of
better and more efficient services. The role of the State Department of
Statistics is also in the process of being upgraded to improve management
and the fulfilment of users needs at various levels.

The Department of Statistics currently produces a wide range of economic

and social statistics, which are compiled based on the priorities and needs
of the government. For example, information on national accounts, balance
of payments, external trade, manufacturing, construction, mining, services,
prices, demography and socio-economics have been published.

Publication is the main channel of data dissemination used by the

department; however, it also entertains usersÊ requests for data in the form
of diskettes, thumb drives, CDs and magnetic tapes. Unpublished statistics
are also available to users on request. Other than that, consultancy services
in the field of sampling are provided by the department to the government
agencies who require them.

(c) Orient Pacific Century (http://www.orientpacific.com) is a specialist

Asian-focused market research and brand research firm assisting both
international and Asian business with business, marketing and brand
strategy. In Malaysia, the Orient Pacific Century has carried out industry
research for the construction and travel and tourism industries and
consumer research for the financial, educational, telecommunication,
 TOPIC 1 INTRODUCTION TO MARKETING RESEARCH  9

aviation, entertainment, software and retail industries. This includes both

qualitative and quantitative methods such as focus groups, street-and-mall
intercepts, telephone surveys (CATI and traditional), and face-to-face and
industry-depth interviews.

The list of studies that have been carried out by Orient Pacific Century in
Malaysia includes the following:

(i) Malaysian Tourism and Promotion Board (MTPB) ă brand positioning

or the international promotion of Malaysian tourism.

(ii) Boral Malaysia for brand image, awareness and competitive brand
positioning.

(iii) Price Waterhouse Coopers: Malaysia, China, Singapore.

(iv) Ipmuda Malaysia.

(v) International Management Centres: Australia, Malaysia, Hong Kong.

(vi) Pacific Forum: Malaysia.

1.5 SELECTING THE RESEARCH SUPPLIER

When initiating a marketing research study, a firm will often need to hire
external suppliers for all or part of the project. The process of selecting an outside
supplier can be formal or informal. First, the hiring firms must develop a
checklist detailing criteria government the selection of outside suppliers. That
checklist should go beyond technical requirements, covering the following areas:

(a) What is the supplierÊs reputation?

(b) Does the firm complete projects on schedule? Is it flexible?

(c) Is it known for maintaining ethical standards?

(d) Are its research projects of high quality?

(e) What kind of and how much experience do the suppliers have?

(f) Does the firm have experience with projects similar to this one?

(g) Do the suppliersÊ personnel have both technical and non-technical

expertise?
10  TOPIC 1 INTRODUCTION TO MARKETING RESEARCH

1.6 CAREERS IN MARKETING RESEARCH

Promising career opportunities are available with marketing research firms (such
as ACNielsen Malaysia, and the Department of Statistics). Equally appealing are
careers in business and non-business firms and agencies with in-house marketing
research departments (e.g., Procter & Gamble, Coca-Cola, LÊOreal Marketing and
the Department of Statistics Malaysia).

A career in research often begins with a supervisory position in field work or

data analysis. With good experience, the researcher may move up to a project
management position. The most common entry-level position in the marketing
research industry is for people with a BachelorÊs degree (for example, BBA). An
operational supervisor is responsible for supervising the day-to-day operations
of specific aspects of the marketing research process. Those with MBA or
equivalent degrees are likely to be employed as project managers. The project
manager works with an account director in managing the day-to-day operations
of a marketing research project.

As indicated by these job titles, careers in marketing research can be either highly
technical, specialising in the design and the statistical aspect of the industry,
or they can be of general management in nature, with emphasis on client
relationships. To prepare for a career in marketing research, you should do the
following:

(a) Take marketing courses offered in your diploma/degree programme;

(b) Take courses in statistics and quantitative methods;

(c) Acquire computer and Internet skills;

(d) Take courses in psychology and consumer behaviour;

(e) Acquire effective written and verbal communication skills; and

(f) Think creatively. Creativity and common sense command a premium in

marketing research.

It is important for marketing researchers to have a broad perspective so that they

can better understand the problems managers face and offer suitable solutions. A
career in marketing research will involve working with other bright, energetic
people in a fast-paced and dynamic environment.
 TOPIC 1 INTRODUCTION TO MARKETING RESEARCH  11

1.7 ECONOMIC SYSTEMS

A marketing information system (MIS) is a structure consisting of people,
equipment and procedures to gather, sort, analyse, evaluate as well as
distribute needed, timely and accurate information to marketing decision
makers. The role of the MIS is to support the decision-making process.

However, this definition sounds very much like that of marketing research,
which is to provide information to aid in decision making. We can better
understand the distinction by examining the components of MIS.

MIS is designed to assess managersÊ information needs, to gather this

information, and to distribute the information to the marketing managers who
need to make decisions. Information is gathered and analysed by the four
subsystems of MIS: internal report system, marketing intelligence, marketing
decision support and marketing research (see Figure 1.2).

Figure 1.2: The marketing information system

Source: Armstrong, G., & Kotler, P. (2010) Marketing: An introduction.
(8th ed.). New Jersey: Pearson Education.
12  TOPIC 1 INTRODUCTION TO MARKETING RESEARCH

ACTIVITY 1.2

In your opinion, what is the importance of MIS to a firm?

1.7.1 The Internal Report System

The Internal Report System gathers information generated by internal reports,

which includes orders, billing, receivables, inventory levels and stock outs.

The internal report system is also commonly referred to as the accounting

information system. Although this system produces financial statements (balance
sheet and income statements, etc.) that generally contain insufficient details for
many marketing decisions, the internal report system also contains details on
both revenues and costs that can be invaluable in making decisions. For example,
a sophisticated internal report system can yield historical information for a given
product, product line, location or region. A good internal report system can tell a
manager a great deal of information about what has happened within the firm in
the past.

ACTIVITY 1.3

Based on your understanding, how does the internal report system

benefit managers?

1.7.2 Marketing Information System (MIS)

MIS includes both informal and formal information-gathering procedures.
Formal information-gathering activities may be conducted by staff members
who are assigned the specific task of looking for any information that seems
significant to the company or industry. Informal information-gathering
procedures involve such activities as scanning newspapers, magazines and trade
publications. This information is then edited and disseminated to the appropriate
individuals or company departments.
 TOPIC 1 INTRODUCTION TO MARKETING RESEARCH  13

1.7.3 Marketing Decision Support System (DSS)

A Marketing Decision Support System is defined as collected data that may be

assessed and analysed using tools and techniques that assist managers in
decision making.

Once companies collect large amounts of information, they store this

information in huge databases that, when assessed with decision-making tools
and techniques (such as break-even analysis, regression models and linear
programming), allow companies to ask „what-if‰ questions. The answers to these
questions are then immediately available for decision making.

1.7.4 Marketing Research Systems

A marketing research system, as discussed and defined earlier, is the fourth
component of MIS. If marketing research and MIS are both designed to provide
information for decision makers, how are the two then different?

(a) First, the marketing research system gathers information not gathered by
other component subsystems. Marketing research studies are conducted
for a specific situation the company is facing. It is unlikely that other
components of an MIS have generated particular information needed for
the specific situation. For example, when „Jelita‰ magazine wants to know
which of the three cover stories it should feature, can its managers obtain
that information from internal reports? The answer is, „No‰. Can this
information be obtained from the intelligence system or the DSS? The
answer is still „No‰. This is how marketing research plays a unique role in
the total information system of the firm. By providing information for a
specific problem at hand, it provides information not provided by other
components of the MIS. This is why those in the industry sometimes refer
to marketing research studies as „ad hoc studies.‰ (Ad hoc in Latin means
„with respect to a specific purpose.‰)

(b) There is another characteristic of marketing research that differentiates it

from the other MIS components. Marketing research projects, unlike the
previous components, are not continuous ă they have a beginning and an
end. This is why marketing research studies are sometimes referred to as
„projects.‰ The other components are available for use only when there is a
justifiable need for information that is not available from internal reports,
intelligence or DSS.
14  TOPIC 1 INTRODUCTION TO MARKETING RESEARCH

SELF-CHECK 1.2

To enhance your understanding, draw a mind map of the four

MIS subsystems.

1.8 ETHICS AND MARKETING RESEARCH

As in most business activities, there exist many opportunities for unethical (and
ethical) behaviour in the marketing research industry.

Ethics is defined as a field of inquiry into determining what behaviours are

deemed appropriate under certain circumstances as prescribed by codes of
behaviour that are set by society. Society determines what is ethical and what
is unethical. In some cases, these standards are formalised by our institutions.

There are many philosophies that may be applied to explain oneÊs determination of
appropriate behaviour given certain circumstances. In the following discussion, we
use two philosophies, deontology and teleology, to explain this behaviour.

Deontology is concerned with the rights of the individual. Is the behaviour fair and
just for each individual? If an individualÊs rights are violated, then the behaviour is
not ethical. For example, consider the marketing research firm that has been hired
to study how customers are attracted to and react to a new form of in-store display.
Here the researchers, hidden from view, record the behaviour of unsuspecting
shoppers as they walk through the supermarkets. A deontologist considers this
form of research activity as unethical because it violates the individual shopperÊs
right to privacy. The deontologist would agree to the research ă if the shoppers
were informed beforehand that their behaviour would be recorded, giving them
the option to participate or not to participate in the study.

On the other hand, teleology analyses a given behaviour in terms of its benefits
and costs to society. If there are individual costs but group benefits, then there
are net gains and the behaviour is considered to be ethical. In our example of the
shopper being observed in the supermarket, the teleologist might conclude that,
 TOPIC 1 INTRODUCTION TO MARKETING RESEARCH  15

although there is a violation in the right to privacy among those shoppers

observed (the cost), there is a benefit if the company learns how to market goods
more efficiently, thus reducing long-term costs. Since this benefit ultimately is
shared by many more individuals than those whose privacy was invaded during
the original study, the teleologist would tend to declare this research practice to
be ethical.

Thus, whether you view the behaviour as being ethical or unethical depends on
your philosophy. Are you a deontologist or teleologist? It is difficult to answer
that question until you are placed in an ethically-sensitive situation. Since you
will certainly come across this situation during your career, you must ask
yourself „How will I respond to such a situation?‰.

Marketing research activities affect four stakeholders:

(a) Marketing researcher;
(b) Client;
(c) Respondent; and
(d) Public.

Ethical questions arise when conflict occurs among these stakeholders.

ACTIVITY 1.4

In your opinion, which philosophy is better, deontology or teleology?

Why?

EXERCISE 1.1

Go to the library and look through several business periodicals such

as Advertising Age, Business Week, Fortune and Forbes. Find three
examples of companies using marketing research.
16  TOPIC 1 INTRODUCTION TO MARKETING RESEARCH

EXERCISE 1.2

Select a company in a field in which you have a career interest and

look up information on this firm in your library or on the Internet.
After having some knowledge of this company, its products and
services, customers and competitors, list five different types of
decisions that you believe this companyÊs management may have
made within the last two years. For each decision, list the information
the companyÊs executives would have needed in order to make these
decisions.

 Marketing research provides information about consumers, channel members,

competition, changes and trends in the marketplace, and other aspects of a
firmÊs environment. It assesses information needs and provides relevant
information in order to improve the marketing decision-making process.

 Marketing research is a systematic and objective process designed to identify

and solve marketing problems. The marketing research process consists of six
steps that must be followed systematically.

 There are two alternatives available for all firms that wish to conduct
marketing research: to conduct the marketing research on its own or to
purchase it from external suppliers. The services offered by external suppliers
range from full-services to specialised services.

 Technological developments are reshaping the role of marketing research as

well as the way in which it is conducted. The Internet can be used at every
step of the marketing research process.

 The ethical issues in marketing research involve four stakeholders, namely

the marketing researcher, the client, the respondent and the public.
 TOPIC 1 INTRODUCTION TO MARKETING RESEARCH  17

Ethics Marketing information system (MIS)

Deontology Marketing research
Internal report system Teleology
Marketing decision support system
(DSS)
Topic  Defining the
Marketing
2 Research
Problem
LEARNING OUTCOMES
By the end of this topic, you should be able to:
1. Describe the importance of and the process used for determining the
marketing research problem;
2. Discuss the tasks involved in problem definition including
discussion with decision maker(s), interview with industry experts,
secondary data analysis, and qualitative research;
3. Evaluate the environmental factors affecting the definition of
research problems; and
4. Summarise the distinction between a management decision problem
and a marketing research problem.

 INTRODUCTION
This topic covers two of the six steps in the marketing research process:

(a) Defining the marketing research problem; and

(b) Developing an approach to the problem.

 TOPIC 2 DEFINING THE MARKETING RESEARCH PROBLEM  19

Defining the problem is the most important step because only when the problem
has been clearly and accurately identified, can a research project be conducted
properly. We will discuss the guidelines for appropriately defining the marketing
research problem and avoiding common types of errors.

2.1 DEFINING THE PROBLEM

Why is it important to define the problem?

A formal research process should not begin until the problem has been clearly
defined. However, properly and completely defining a marketing problem is
easier said than done. When the problem or opportunity is discovered, marketing
managers may have only vague insights into a complex situation. For example,
suppose market share is declining on the East Coast and management does
not know the reason for the drop. If quantitative research was initiated before
the manager learns exactly what is important, the investigation may yield
false conclusions. The right answer to the wrong question may be absolutely
worthless-worse, it may even be harmful.

Problem definition is a broad statement of the general problem and

identification of the specific component of the marketing research problems. It
also specifies the research questions to be answered and the objectives of the
research.

A clearly defined problem serves as a guideline to the researcher in designing

and conducting research properly. Thus, it helps the researcher to answer the
question: „What is to be done?‰.

A clearly laid down research problem leads to goal-directed research, which will
meet the objectives of the decision maker. This research is better than haphazard
research which often results in incomplete information.

ACTIVITY 2.1

In your opinion, what will happen if a firm fails to clearly define

a problem?
20  TOPIC 2 DEFINING THE MARKETING RESEARCH PROBLEM

2.2 THE PROCESS OF DEFINING THE PROBLEM

AND DEVELOPING AN APPROACH
Formulating the problem is a sequential process which comprises the following
steps:

(a) Discussion with the decision maker;

(b) Interview with industry experts;

(c) Secondary data analysis; and

(d) Translating the marketing decision problem into a marketing research

problem.

2.2.1 Discussion with the Decision Maker

The researcher must communicate with the decision maker in order to
understand the nature of the problems the decision maker faces and what
information the research could provide. Such an understanding will help the
researcher gather information relevant to the problem faced by the decision
maker. Note that a candid and open discussion between the researcher and the
decision maker may help in identifying:

(a) Events that led to the need for decision making;

(b) Alternative courses of action available to decision makers;

(c) Criteria to be used in evaluating various courses of action; and

(d) Information that is needed by the decision maker.

A systematic approach to working with the decision makers is called the

„problem audit‰. It enables the researcher to move beyond the symptoms to
understand the causes of the problem.

Access to decision makers may be difficult as they may not have enough time for
a fruitful discussion with the researcher. This is especially true in the case of
multiple decision makers, where there may be conflicts of both interest and time.
Sometimes the decision makers decide on the action in advance and simply want
the data that will support their plans. This is not a sound application of
marketing research and the researcher should remain unbiased while conducting
and presenting a study.
 TOPIC 2 DEFINING THE MARKETING RESEARCH PROBLEM  21

2.2.2 Interview with Industry Experts

Industry experts can provide useful information about prevailing market
conditions. They can be especially useful in the case of industrial marketing
research where technical knowledge is required. Experts can be found both
inside and outside the firm.

The purpose of interviewing experts is to help define the marketing problem

rather than to develop a conclusive solution. Unfortunately, two potential
difficulties may arise when seeking advice from experts:
(a) Some individuals who claim to be knowledgeable and are eager to
participate may not really possess the required expertise; and
(b) It may be difficult to locate and obtain help from experts who are outside
the clientÊs organisation.

An example of an industry expert in Malaysia is the Malaysian Automotive

Association (MAA), which specialises in the motor industry. MAA will assist and
advise any members who may be engaged in enforcing any principle considered
to be for the benefit of the motor industry or in resisting any action or proposal
considered by the association to be against the best interests of the motor
industry. In addition, the association is also responsible for communicating with
the public through various media on industry issues, positions and objectives.

2.2.3 Secondary Data Analysis

Primary data is data that has been obtained by the researcher for the specific
purpose of addressing a research problem. Secondary data, on the other hand, is
data collected for some purpose other than the problem at hand. Analysis of
available secondary data is an essential step in the problem definition process.
Primary data should not be collected until all available secondary data has been
fully analysed.

2.2.4 Translating the Marketing Decision Problem

into a Marketing Research Problem
It is important that the marketing decision problem be translated into a
marketing research problem. The marketing research problem must be
significant in nature and be specific, measurable, achievable, and realistic and
have a time-frame for completion. A detailed analysis on the marketing research
problem is discussed in Section 2.4.
22  TOPIC 2 DEFINING THE MARKETING RESEARCH PROBLEM

SELF-CHECK 2.1

Discuss the process of defining the problem.

2.3 THE NATURE OF THE PROBLEM

The nature of the problem consists of the factors that have an impact on the
definition of the marketing research problem, including past information and
forecasts, resources and constraints of the firm, objectives of the decision maker,
buyer behaviour, legal environment, economic environment, and marketing and
technological skills of the firm. Each of these is discussed below:

(a) Past Information and Forecasts

They provide industry data, which puts the current problem into context.

(b) Resources and Constraints

They force the problem to be defined within an appropriate scope.

(c) Objectives of the Decision Maker

The direction in which the research will be conducted will depend upon the
objectives of the decision maker.

(d) Buyer Behaviour

An understanding of consumer behaviour is critical to understand how
consumers will respond to specific marketing actions.

(e) Legal Environment

The legal environment may regulate certain aspects of the marketing mix
and the research effort, thus affecting the problem definition.

(f) Economic Environment

This can affect the decisions of consumers and impact of the marketing mix.

(g) Marketing and Technological Skills

The abilities of the organisations to develop and market products may
affect the scope of the research to be conducted. In addition, technological
advances will determine the methods of conducting marketing research.

After gaining an adequate understanding of the environmental context of the

problem, the researcher can then define the management problem and the
marketing research problem.
 TOPIC 2 DEFINING THE MARKETING RESEARCH PROBLEM  23

2.4 MARKETING DECISION PROBLEMS AND

MARKETING RESEARCH PROBLEMS
The marketing decision problem defines what the decision maker needs to do,
whereas the marketing research problem entails determining what information
is needed and how it can be obtained in the most feasible way. To show the
differences between the two types of problems, let us see the following example
in Table 2.1.

Table 2.1: Differences between Marketing Decision Problem and

Marketing Research Problem

Marketing Decision Problem Marketing Research Problem

1. Should the price be cut in response to a 1. Determine the buyer behaviour at
price cut by a competitor? various price levels.
2. Should „Keropok Segera‰ be 2. Assess the probable market size and
introduced in the market? market share for Keropok Segera.
3. What should be done to increase the 3. Determine the relative strengths and
market share of „Kicap Tamin‰? weaknesses of Kicap Tamin vis-a-vis
those of the competitors.

Source: Malhotra, N. (2002). Basic marketing research: Applications to

contemporary issues. Upper Saddle River, New Jersey, USA: Prentice-Hall, Inc.

2.5 DEFINING THE MARKETING RESEARCH

PROBLEM
A well-defined marketing research problem consists of both a broad statement
and a list of specific components. The broad statement provides a perspective
on the problem and acts as a safeguard against overlooking important aspects
of the marketing research and management decision problems. The specific
components focus on the key aspects of the problem and provide clear guidelines
on how to proceed further.

A broad definition does not provide guidelines for subsequent steps in research.
A narrow definition, on the other hand, may preclude the consideration of some
courses of action. In either case, the solution that is reached will not be directly
related to the problem and may be of little use to the manager.
24  TOPIC 2 DEFINING THE MARKETING RESEARCH PROBLEM

2.6 COMPONENTS OF APPROACH TO

RESEARCH
The following are components of the approach to research.

(a) Objective/Theoretical Framework

In general, research should be based on objective evidence and supported
by a theory. A theory is a conceptual scheme based on foundational
statements called axioms, which are assumed to be true. Objective evidence
(evidence that is unbiased and supported by empirical findings) is gathered
by compiling relevant findings from secondary sources.

An appropriate theory to guide the research might be identified by

reviewing academic literature contained in books, journals and monographs.
Theoretical considerations provide information on how the variables should
be selected. A theory also serves as a foundation on which the researcher
can organise and interpret his findings.

(b) Analytical Model

An analytical model is a set of variables and their inter-relationships
designed to represent, in whole or in part, some real system or process. The
model may have different forms such as verbal model, graphic model and
mathematical model (see Table 2.2).

Table 2.2: Analytical Models

Forms of Analytical
Model Description

Verbal Model  Provides a written representation of the relationship

between variables.
 Often a summary or restatement of the main points of
theory.
Graphic Model  Provides a visual picture of the relationships between
variables.
 Used to isolate variables and to suggest direction of
relationships.
 Is not designed to provide numerical results.
Mathematical Model  Describes the relationship between variables, usually as an
equation.
 TOPIC 2 DEFINING THE MARKETING RESEARCH PROBLEM  25

Graphical models are particularly helpful in conceptualising an approach

to the problem, as the following jeans purchase model illustrates.

LEE RIVETED RIVETS YOUNG CONSUMERS WITH “THE BRANDS

THAT FITS”
According to consumer decision-making theory, the consumer first decides
whether to purchase jeans or other casual clothes. If jeans are to be
purchased, the consumer will form selection criteria for evaluating
alternative brands. The selection criteria consist of factors such as color,
price, fit, cut, comfort, and quality. The competing brands of jeans are then
evaluated based on the selection criteria to purchase one or more brands.
The following graphical model illustrates the decision process for a
consumer considering the purchase of casual clothing. Lee Riveted jeans
are target at young people who buy jeans based primarily on fit and cut.
Therefore, the marketing themes of Lee Riveted with tag lines such as “the
brand that fits,” “cut to be noticed,” and “that unstoppable spirit” were based
on this model.

Source: Malhotra, N. (2002). Basic marketing research: Applications to contemporary

issues. Upper Saddle River, New Jersey, USA: Prentice-Hall, Inc.
26  TOPIC 2 DEFINING THE MARKETING RESEARCH PROBLEM

The verbal, graphical and mathematical models complement each other and
help the researcher identify relevant research questions and hypotheses, as
shown in Figure 2.1.

Figure 2.1: Development of research questions and hypothesis

Source: Malhotra, N. (2002). Basic marketing research: Applications to contemporary
issues. Upper Saddle River, New Jersey, USA: Prentice-Hall, Inc.

(c) Research Question

What are research questions?

Research questions are refined statements of the specific components of

the research problems. Research questions ask what specific information is
required with respect to the problem components. The formulation of the
research question should be guided not only by the problem definition, but
also by the theoretical framework and the analytical model adopted.

For instance, a personal computer company made the following statement

about an advertising problem: „In the broadest sense, the marketing
problem is to determine the best ways (our company) can communicate
with potential buyers of laptop computers.‰

This leads to the following research questions:

(i) How familiar are consumers with the various brands of computers?

(ii) What attitudes do consumers have towards brands of computers?

 TOPIC 2 DEFINING THE MARKETING RESEARCH PROBLEM  27

(iii) How important are the various factors for evaluating the purchase of
a laptop computer?

(iv) How effective are the communication efforts of the various competing
marketers in terms of message recognition?

Research questions make it easier to understand the issues that are

confusing managers and to indicate what issues have to be resolved. A
research question is the researcherÊs translation of the marketing problem
into specific inquiry.

A research question can be too vague and general if stated in terms such
as „Is advertising copy X better than advertising copy Y‰. Advertising
effectiveness can be measured by sales, recall-of-sales message, brand
awareness, intention to buy and so on. Asking a more specific research
question (e.g. „Which advertisement has a higher day-after recall score?‰)
helps the researcher design a study that will produce pertinent information.
The answer to the research question should be a criterion that can be used
as a standard for selecting alternatives. The stage of research obviously is
related to problem definition. The goal of defining the problem is to state
the research question clearly and to develop a well-formulated hypothesis.

(d) Hypothesis
What is a hypothesis?

A hypothesis is an unproven statement or proposition about a factor or

phenomenon that is of interest to the researcher. Hence, hypotheses go
beyond research questions because they are statements of relationships or
propositions rather than merely questions to which answers are sought.
Unfortunately, it may not be possible to formulate hypotheses in all
situations. Sometimes sufficient information is not available to develop
hypotheses. At other times, the most reasonable statement of a hypothesis
may be a trivial restatement of the research question.

For instance, a sales manager may hypothesise that salespeople who

show the highest job satisfaction are the most productive. An advertising
manager may believe that if consumersÊ attitudes towards a product are
changed in a positive direction, consumption of the product will increase.
Problem statements and hypotheses are similar because both state
relationships. However, problem statements are interrogative, whereas
hypotheses are declarative.
28  TOPIC 2 DEFINING THE MARKETING RESEARCH PROBLEM

A formal statement of a hypothesis has considerable practical value in

planning and designing research. It forces researchers to be clear about
what they expect to find through the study, and it raises crucial questions
about the data that will be required in the analysis stage. When evaluating a
hypothesis, a researcher should ensure that the information collected will
be useful in decision making.

Notice how the following hypotheses express expected relationships

between variables:

(i) There is a positive relationship between buying on the Internet and

the presence of younger children in the home.

(ii) Sales are lower for salespeople in regions that receive less advertising
support.

(iii) Consumers will experience cognitive dissonance after the decision to

purchase a Sony personal video recorder.

(iv) Opinion leaders are more affected by mass media communication

sources as compared to non-leaders.

(e) Specification of Information Needed

By focusing on each component of the problem as well as on the analytical
framework, models, research questions, and hypotheses, the researcher can
determine what information should be obtained in the marketing research
project. It is helpful to carry out this exercise for each component of the
problem and make a list specifying all information that should be collected.

EXERCISE 2.1

State a problem in your field of interest and list some variables that
might be investigated to solve this problem.

Go to the library, find some business journals, and record and

evaluate some hypotheses that have been investigated in recent years.
Identify the key independent and dependent variables.
 TOPIC 2 DEFINING THE MARKETING RESEARCH PROBLEM  29

EXERCISE 2.2

Evaluate the following statements of marketing problems:

(a) The producer of a television show: „We have a marketing

problem. The programme rating is low. We need to learn how
we can improve our ratings.‰

(b) A soft drink manufacturer: „The marketing problem is that we

do not know if our bottlers are more satisfied with us than our
competitorÊs bottlers are with them.‰

(c) A farm implement manufacturer: „Our objective is to learn

the most effective form of advertising so we can maximise
product line profits.‰

 Defining the marketing research problem is the most important step in

a marketing research project. It is a difficult step because management
frequently has not determined the actual problem or has only a vague idea
about it. The researcherÊs role is to help management identify and isolate the
problem.

 The tasks involved in formulating the marketing research problem include

discussion with management, including the key decision makers; interviews
with industry experts; analysis of secondary data and qualitative research.

 These tasks should lead to an understanding of the environmental context of

the problem.

 Analysis of the environmental context should assist in the identification of

the management decision problem, which should then be translated into a
marketing research problem.

 The researcher should avoid defining the marketing research either too
broadly or too narrowly.

 An appropriate way of defining the marketing research problem is to make a

broad statement of the problem and then identify its specific components.
30  TOPIC 2 DEFINING THE MARKETING RESEARCH PROBLEM

Analytical model Research question

Hypothesis Secondary data analysis
Industry experts Theoretical framework
Qualitative research
Topic  Research
Design
3
LEARNING OUTCOMES
By the end of this topic, you should be able to:
1. Explain two types of research design;
2. Compare the appropriateness of the three basic types of research
design: exploratory, descriptive and causal;
3. Assess the major errors in a research design;
4. Analyse the elements of a marketing research project and show how
it addresses the steps of the marketing research process;
5. Elaborate on how information technology can be used to improve
components of research design and minimise research error; and
6. Propose a research design.

 INTRODUCTION
This topic defines and classifies research design. We will describe two major
types of research design: exploratory and conclusive (which includes both
descriptive and causal research design). We will also consider the differences
between two types of descriptive design, cross-sectional and longitudinal, and
identify the various types of common errors.
32  TOPIC 3 RESEARCH DESIGN

3.1 WHAT IS RESEARCH DESIGN?

Do you know what the purpose of research design is?

Research design is a framework or blueprint for conducting the marketing

research project. It specifies the procedures necessary to obtain the information
needed to structure and/or solve marketing research problems.

The research design lays the foundation for conducting a project. A good
research design will ensure that the marketing research project is conducted
effectively and efficiently. The process of designing the research study involves
many interrelated decisions. The most significant decision is the choice of
research approach because it determines how the information will be obtained.
Typical questions at this stage are:

(a) Should we rely on secondary sources such as the Census?

(b) Which is more appropriate: an exploratory approach with group

discussions or a survey?

(c) Is a mail, telephone, fax or personal interview survey better for this
problem?

Tactical research decisions are made once the research approach has been chosen.
Here, the focus is on the specific measurement to be made or questions to be
asked, the structure and length of the questionnaire, and the procedure for
choosing a sample to be used. These tactical decisions are also constrained by
time and budget availability. Therefore, before a study can be implemented, the
estimated cost must be compared with the anticipated value.

To design something also means to ensure that the pieces fit together. The
achievement of this fit among objectives, research approaches and research
tactics is inherently an iterative process in which earlier decisions are constantly
reconsidered in light of subsequent decisions. This may mean a revision of the
research objectives as new insights are gained into the complexities of the
population to be sampled, or a reassessment of the research approach in light of
realistic cost estimates.
 TOPIC 3 RESEARCH DESIGN  33

3.2 BASIC RESEARCH DESIGNS

Research designs are of two broad types: exploratory and conclusive. Conclusive
research design may be further categorised as either descriptive or causal.
Descriptive design may be further categorised as either cross-sectional or
longitudinal (see Figure 3.1).

Figure 3.1: Classification of market research design

Source: Malholra, N. (2002). Basic marketing research: Applications to contemporary
issues. Upper Saddle River, New Jersey, USA: Prentice-Hall.

3.2.1 Exploratory Research

Exploratory research is conducted to clarify ambiguous problems. The primary
objective of exploratory research is to explore the problem situation, that is,
to gain ideas and insights into the problem confronting management or
the researcher. It is not intended to provide conclusive evidence from which
to determine a particular course of action. Exploratory research is usually
conducted with the expectation that subsequent research will be required to
provide such conclusive evidence.

Exploratory research could be used for any of the following purposes:

(a) Formulate a problem or define a problem more precisely;
(b) Identify alternative courses of action;
(c) Develop hypotheses;
(d) Isolate key variables and relationships for further examination;
34  TOPIC 3 RESEARCH DESIGN

(e) Gain insights for developing an approach to the problem; and

(f) Establish priorities for further research.

For example, suppose a Chinese fast-food restaurant chain is considering

expanding its hours and product line with a breakfast menu. Exploratory
research with a small number of current customers might identify a strong
negative reaction to eating spicy vegetable breakfast at a Chinese fast-food outlet.
Thus, exploratory research might help crystallise a problem and identify
information needed for future research.

A variety of methods which are available to conduct exploratory research

include:
(a) Secondary data analysis;
(b) Experience survey;
(c) Case analysis;
(d) Focus groups; and
(e) Projective techniques

which will be discussed in greater detail in Topic 4.

Table 3.1 shows the differences between exploratory and conclusive research.

Table 3.1: Differences between Exploratory and Conclusive Research

Aspect Exploratory Conclusive

Objectives To provide insights and To test specific hypotheses and
understanding examine relationships
Characteristics  Information needed is  Information needed is clearly
defined only loosely defined
 Research process is flexible  Research process is formal
and unstructured and structured
 Sample is small and non-  Sample is large and
representative representative
 Analysis of primary data is  Data analysis is quantitative
Qualitative
Findings/Results Tentative Conclusive
Outcome Generally followed by further Finding used as input to
exploratory or conclusive decision making
research

Source: Malhotra, N. (2002). Basic marketing research: Applications to contemporary

issues. Upper Saddle River, New Jersey, USA: Prentice-Hall, Inc.
 TOPIC 3 RESEARCH DESIGN  35

3.2.2 Conclusive Research

Conclusive research is designed to assist the decision maker in determining,
evaluating and selecting the best course of action in a given situation. Conclusive
research can also be used to verify the insights gained from exploratory research.

(a) Descriptive Research

As the name implies, the main purpose of this research is to describe the
characteristics of some complex marketing phenomena.

Marketing managers frequently need to determine who purchases a

product, who buys a specific brand, what the size of the market is, what the
competitorsÊ actions are and so on. Hence, descriptive research seeks to
determine the answers to who, what, when, where and how questions.

Unlike exploratory research, descriptive studies are based on some

previous understanding of the nature of the research problem. Although
the researcher may have a general understanding of the situation, the
conclusive evidence that answers questions of fact necessary to determine a
course of action has yet to be collected. Many circumstances require
descriptive research to identify the reasons consumers give to explain the
nature of things. In other words, a diagnostic analysis is performed when
consumers are asked questions such as „Why do you feel that way?‰
Although they may describe why consumers feel a certain way about a
product or situation, the findings of a descriptive study such as this,
sometimes called diagnosis, do not provide causal evidence. Frequently,
descriptive research attempts to determine the extent of differences in
needs, attitudes and opinions among subgroups.

In short, descriptive research is conducted for the following reasons:

(i) To describe the characteristics of relevant groups, such as consumers,

salespeople, organisations, or market areas. For example, we could
develop a profile for „heavy users‰ (frequent shoppers) of Suria
KLCC.

(ii) To estimate the percentage of units in a specified population

exhibiting a certain behaviour. For example, we might be interested in
estimating the percentage of heavy users of prestigious department
stores who also patronise discount department stores.
36  TOPIC 3 RESEARCH DESIGN

(iii) To determine the perception of product characteristics. For example,

how do households perceive the various department stores in terms
of salient factors of the choice criteria?

(iv) To determine the degree to which marketing variables are associated.

For example, to what extent is shopping at department stores related
to eating out?

(v) To make specific predictions. For example, what will the retail sale of
Jaya Jusco for fashion clothing in the Petaling Jaya area be?

Descriptive research can be further classified into cross-sectional and

longitudinal research.

(i) Cross-sectional Design

A cross-sectional design involves the collection of data from a sample
at a single point in time. For example, if a researcher observes
the buyer behaviour from randomly chosen people in a mall and
then analyses the data gathered, such a study will be called a
cross-sectional study. Cross-sectional design has an advantage in
representative sampling and response bias. Multiple cross-sectional
design, on the other hand, involves collecting data from two or more
samples of respondents and information from each sample.

(ii) Longitudinal Design

A longitudinal design involves collecting data periodically from a
fixed sample of respondents. Thus, a longitudinal study provides
a series of pictures that, when viewed together, portray a vivid
illustration of the situation and the changes that are taking place.
Compared to cross-sectional design, longitudinal design is more
effective in detecting change, more accurate, and provides a larger
amount of data. Sometimes, the term „panel‰ is used interchangeably
with the term „longitudinal.‰ A panel consists of a sample of
respondents, generally households that have agreed to provide
information at specified intervals over an extended period. Syndicated
firms maintain panels, and panel members are compensated for their
participation with gifts, coupons, information or cash.

Data obtained from panels not only provide information on market

shares that are based on an extended period of time, but also allow the
researcher to examine changes in market share over time.
 TOPIC 3 RESEARCH DESIGN  37

(b) Causal Research

The main goal of causal research is to identify cause-and-effect relationships
among variables. Exploratory and descriptive research normally precede
cause-and-effect relationship studies. In causal studies, researchers typically
have an expectation about the relationship to be explained such as a
prediction about the influence of price, packaging, advertising, and the like
on sales. Thus, a researcher must be quite knowledgeable about the subject.

Ideally, the manager wants to establish that one event (say, new packaging)
is the means for producing another event (an increase in sales). Causal
research attempts to establish that when we do one thing, another thing
will follow. The word „cause‰ is common in everyday conversation, but
from a scientific research perspective, a true causal relationship is impossible
to prove. Nevertheless, researchers seek certain types of evidence to help
them understand and predict relationships.

The comparison of basic research designs is shown in Table 3.2.

Table 3.2: Comparison of Basic Research Designs

Aspect Exploratory Descriptive Causal

Objectives Discover ideas and Describe market Determine cause-and-
insights. characteristics or effect relationship.
functions.
Characteristics  Flexible  Marked by the  Manipulation of
 Versatile prior formulation one or more
 Often the front end of specific independent
of total research hypotheses variables
design  Pre-planned and  Control of other
structured design mediating
variables
Methods  Expert surveys  Secondary data  Experiments
 Pilot surveys  Surveys
 Case study  Panels
 Secondary data  Observational and
 Qualitative other data
Research

Source: Malhotra, N. (2002). Basic marketing research: Applications to contemporary

issues. Upper Saddle River, New Jersey, USA: Prentice-Hall, Inc.
38  TOPIC 3 RESEARCH DESIGN

Table 3.3, on the other hand, illustrates the application of the three research
approaches.

Table 3.3: Three Research Approaches

Research Purpose Research Questions Hypothesis

Exploratory Research
1. What new product What alternative ways are Boxed lunches are better
should be developed? there to provide lunches for than other forms.
school children?
2. What product appeal What benefits do people Constructs unknown.
will be effective in seek from the product?
advertising?
3. How can our service be What is the nature of any Suspect that an image of
improved? customer dissatisfaction? impersonalisation is a
problem.
Descriptive Research
1. How should a new Where do people now buy Upper-class buyers use
product be distributed? similar products? specialty stores, and
middle- class buyers use
department stores.
2. What should be the What kinds of people now Older people buy our
target segment? buy the product, and who brands, whereas the young
buys our brands? married are heavy users of
competing brands.
3. How should our What is our current image? We are regarded as being
product be changed? conservative and behind
the times.
 TOPIC 3 RESEARCH DESIGN  39

Causal Research
1. Will an increase in the What is the relationship For small organisations, an
service staff be between size of service staff increase of 50% or less will
profitable? and revenue? generate marginal revenue
in excess of marginal costs.
2. Which advertising What would get people out Advertising programme A
programme for public of cars and into public generates more new riders
transit should be run? transit? than programme B.
3. Should a new budget or Will the „no frills‰ airline The new airfare will attract
„no frills‰ class or generate sufficient new sufficient revenue from new
airfare be introduced? passengers to offset the loss passengers.
of revenue from existing
passengers who switch
from economy class?

Source: Aaker, D. A., Kumar, V., & Day, G. (2001). Marketing research, (7th ed.).
New York, John Wiley & Sons, Inc.

One of the methods of causal research is experimentation. Experiments can

take place either in a laboratory or in a natural setting. A typical causal study
has management change one variable (for example, advertising) and then
observe its effect on another variable (such as sales). Some evidence for
causality comes from the fact that the cause precedes the effect. In other words,
having an appropriate causal order of events, or temporal sequence, is one
criterion for causality that must be met to be able to establish a causal
relationship. If a consumer behaviour theorist wishes to show that an attitude
change causes a change in behaviour, one criterion that must be established is
that attitude change precedes the behaviour change.

SELF-CHECK 3.1

Discuss the three research approaches.

40  TOPIC 3 RESEARCH DESIGN

3.3 RELATIONSHIPS AMONG EXPLORATORY,

DESCRIPTIVE AND CAUSAL RESEARCH
The relationship among exploratory, descriptive and causal research is that if not
much is known about the problem situation, it might be desirable to begin with
exploratory research.

However, exploratory research should not be used by itself; rather, exploratory

research should be followed by descriptive and/or causal research.

Depending on the precision with which the problem has been defined and
the degree of understanding the researcher has with respect to developing an
approach to the problem, a research design could well begin with descriptive or
causal research and, if possible, be followed by exploratory research.

These are the guidelines for choosing a research design:

(a) When little is known about the problem situation, it is good to begin
with exploratory research. Exploratory research is appropriate when the
problem needs to be defined more precisely, alternative courses of action
identified, research questions or hypotheses developed, and key variables
isolated and classified as dependent or independent.

(b) Exploratory research is the initial step in the overall research design
framework. It should, in most instances, be followed by descriptive or
causal research.

(c) It is not necessary to begin every research design with exploratory research.
It depends upon the precision with which the problem has been defined
and the researcherÊs degree of certainty about the approach to the problem.
A research design could well begin with descriptive or causal research.

(d) Although exploratory research is generally the initial step, it need not
always be the case. Exploratory research may follow descriptive or causal
research. For example, descriptive or causal research results in findings that
are hard for managers to interpret. Exploratory research may provide more
insights to help understand the findings.
 TOPIC 3 RESEARCH DESIGN  41

3.4 TASKS INVOLVED IN FORMULATING A

RESEARCH DESIGN
What are the tasks involved in formulating a research design?

In formulating a research design, typically the researcher must perform the

following tasks:

(a) Specify the information needed;

(b) Design the exploratory, descriptive, and/or causal phases of the research
design;

(c) Specify the measurement and scaling procedures;

(d) Construct and pre-test a questionnaire (interview form) or an appropriate

form for data collection;

(e) Specify the sampling process and sample size; and

(f) Develop a plan of data analysis.

3.5 POTENTIAL SOURCES OF ERROR

What are potential sources of error?

Several potential sources of errors can affect a research design. A good research
design attempts to control the various sources of error. The total error is the
variation between the true mean value in the population of variable of interest
and the observed mean value obtained in the marketing research project. In
general, there are two types of errors:
(a) Sampling error; and
(b) Non-sampling error.
42  TOPIC 3 RESEARCH DESIGN

Each of these errors will be discussed in detail. See Figure 3.2.

Figure 3.2: Errors in research design

Source: Aaker, D. A. , Kumar, V. & Day, G., (2009). Marketing research, (10th ed.). New
York, John Wiley & Sons, Inc.

3.5.1 Sampling Error

Sampling error is the difference between a measure obtained from a sample
representing the population and the true measure that can be obtained only
from the entire population. This error occurs because no sample is a perfect
representation of a given population, unless the sample size equals the
population.
 TOPIC 3 RESEARCH DESIGN  43

3.5.2 Non-sampling Error

Non-sampling error includes all other errors associated with a research project.
There may be several different reasons for these errors, which can be broadly
classified into four groups as illustrated in Figure 3.3.

Figure 3.3: Non-sampling error

(a) Design Errors ă also called researcher-included errors, are mainly due to
flaws in the research design. There are several different types of design
errors.

(i) Selection Error ă occurs when a sample obtained through a non-

probability sampling method is not representative of the population.
For example, if a mall interviewer interested in the shopping habits of
visitors to the mall avoids interviewing people with children, he is
inducing a selection error into the research study.

(ii) Population Specification Error ă occurs when an inappropriate

population is chosen from which to obtain data for a research study.
For example, if the objective of a research study is to determine what
brand of dog food people buy for their pets and the research draws a
sample from a population that consists predominantly of cat owners,
a population specification error is induced into the study.

(iii) Sampling Frame Error ă a sampling frame is a directory of population

members from which a sample is selected. A sampling frame error
occurs when the sample is drawn from an inaccurate sampling frame.
For example, if a researcher is interested in finding the reasons why
some people have personal computers in their homes and selects the
sample from a list of subscribers of PC World, he is inducing a sample
frame error into the study.
44  TOPIC 3 RESEARCH DESIGN

(iv) Surrogate Information Error ă is the difference or variation between

the information required for a marketing research study and the
information being sought by the researcher. The famous (or rather
infamous) New Coke taste test is a classic example of surrogate
information error. The researchers in this case were seeking
information regarding the taste of New Coke versus Old Coke, but the
study should have determined consumerÊs attitudes toward a change
in the product and not just their taste preferences.

(v) Measurement Error ă the difference or the variation between the

information sought by a researcher for a study and the information
generated by a particular measurement procedure employed by
the researcher. Measurement error can occur at any stage of the
measurement process, from the development of an instrument to the
data analysis and interpretation stage. For example, if a researcher,
interested in the „individual income‰ of the respondent, words the
question as „annual household income‰, a measurement error is being
induced in the research study.

(vi) Experimental Error ă An experiment is designed to determine the

existence of any causal relationship between two variables. Any error
caused by the improper design of the experiment induces an
experimental error into the study.

(vii) Data Analysis Error ă can occur when the data from the questionnaire
is coded, edited, analysed or interpreted. For example, incorrect
coding of data or wrong use of statistical analysis procedures can
induce a data analysis error in the study.

(b) Administering Errors ă all errors that occur during the administration of a
survey instrument to the respondent are classified as administering errors.
They are caused by mistakes committed by the person administering the
questionnaire. They may be caused by three major factors:

(i) Questioning Error ă arises from improperly-worded questions asked

of the respondent. If the interviewer does not word the question
exactly as designed by the researcher, a questioning error is induced.

(ii) Recording Error ă occurs from improper recording of the respondentÊs

answers. If the interviewer misinterprets the response or hears it
inaccurately, this induces a recording error into the study.
 TOPIC 3 RESEARCH DESIGN  45

(iii) Interference Error ă occurs when an interviewer interferes with, or

fails to follow, the exact procedure while collecting data. For example,
if the interviewer fabricates the responses to a survey, it induces an
interference error.

(c) Response Errors ă also called data errors, occur when the respondent ă
intentionally or unintentionally ă provides inaccurate answers to the survey
questions. This might be due to the respondentÊs failing to comprehend the
question or it may be due to fatigue, boredom or misinterpretation of
the question. Response errors can also occur when a respondent who is
unwilling or embarrassed to answer a sensitive question provides an
inaccurate or false response.

(d) Non-response Errors ă occur if:

(i) Some members of a sample were not contacted, and hence their
responses were not included in the study; or

(ii) Some of the members contacted provide an incomplete response or

did not respond to the survey instrument.

The primary reasons for this error occurring include the unwillingness of a
respondent to participate in the study and the inability of the interviewer to
contact the respondents.

3.6 BUDGETING AND SCHEDULING THE

PROJECT
Budgeting and scheduling help to ensure that the marketing research project is
completed within the available resources: financial, time, manpower and others.
By specifying the time parameters within which each task should be completed
and the costs of each task, the research project can be effectively managed.

3.7 MARKETING RESEARCH PROPOSAL

A marketing research proposal is the official layout of the planned marketing
research activity for management. It describes the research problem, the
approach, the research design, data collection methods, data analysis methods
and reporting methods.
46  TOPIC 3 RESEARCH DESIGN

Although the format of a research proposal may vary considerably, most

proposals address all steps of the marketing research process and contain the
following elements:

(a) Executive Summary ă The proposal should begin with a summary of the
major points from each of the other sections, presenting an overview of the
entire proposal.

(b) Background ă The background of the problem, including the environmental

context, should be discussed.

(c) Problem Definition/Objectives of the Research ă Normally, a statement of

the problem, including the specific components, should be presented. If this
statement has not been developed, the objectives of the marketing research
project should be clearly specified.

(d) Theoretical Framework ă This section explains the theories to be used to

explain the problem at hand.

(e) Literature Review ă At a minimum, a review of the relevant academic and

trade literature should be presented along with some kind of an analytical
model. If research questions and hypotheses have been identified, then
these should be included in the proposal.

(f) The model ă The model is a description of the relationship between the
dependent and independent variables.

(g) Research Design ă The research design adopted, whether exploratory,

descriptive, or causal, should be specified. Information should be provided
on the following components:
(i) Kinds of information to be obtained;
(ii) Method of administering the questionnaire;
(iii) Scaling techniques;
(iv) Nature of the questionnaire; and
(v) Sampling plan and sample size.
 TOPIC 3 RESEARCH DESIGN  47

(h) Field Work/Data Collection ă The proposal should discuss how the data
will be collected and who will collect it.

(i) Data Analysis ă The kinds of data analysis that will be conducted (simple
cross tabulation, univariate analysis and multivariate analysis) and how the
results will be interpreted should be described.

(j) Reporting ă The proposal should specify whether intermediate reports will
be presented and at what stages, what will be the form of the final reports,
and whether a formal presentation of the results will be made.

(k) Cost and Time ă The cost of the project and a time schedule, broken down
by phases, should be presented.

(l) Bibliography ă This section should contain the list of references (books and
journals, etc.) that are used in the research.

(m) Appendices ă Any statistical or other information that is of interest to only

a few people should be contained in appendices.

Preparing a research proposal ensures that the researcher and management agree
about the nature of the project, and it also helps sell the project to management.
As the preparation of the proposal entails planning, it helps the researcher
conceptualise and execute the marketing research project.

EXERCISE 3.1

1. In a short essay, discuss the relationship among exploratory,

descriptive, and causal research and provide some basic guidelines
discussing the appropriate situation in which to apply each type of
research design. Include a specific example of each research design
type to support your answer.
2. In a short essay, list and discuss at least seven elements that are
included in a proposal presenting a detailed research design.
3. In a short essay, list the six tasks that a researcher must typically
perform in formulating a research design.
48  TOPIC 3 RESEARCH DESIGN

 A research design is a framework or blueprint for conducting the marketing

research project. It specifies the detail of how the project should be conducted.
Research design may be broadly classified as exploratory or conclusive.

 The primary purpose of exploratory research is to provide insights into the

problem. Conclusive research is conducted to test specific hypotheses and
examine specific relationships.

 Conclusive research may be either descriptive or causal.

 The major objective of descriptive research is to describe market characteristics

or functions. A descriptive design requires clear specifications of the „who‰,
„what‰, „when‰, „where‰, „why‰ and „how‰ of the research.

 Descriptive research can be further classified into cross-sectional and

longitudinal research.

 Cross-sectional design involves the collection of information from a sample of

population elements at a single point in time.

 In contrast, longitudinal design involves repeated measurements taken on a

fixed sample.

 Causal research is designed for the primary purpose of obtaining evidence

about cause-and-effect (causal) relationships.

Causal research Descriptive research

Conclusive research Exploratory research
Cross-sectional design Longitudinal design
Topic  Data Collection
Method and
4 Sample Design
LEARNING OUTCOMES
By the end of this topic, you should be able to:
1. Distinguish secondary data from primary data;
2. Analyse the advantages and disadvantages of secondary data and
their uses in the various steps of the marketing research process;
3. Classify survey methods;
4. Discuss the criteria for selecting survey methods and the various
methods to improve survey response rate;
5. Review the relative advantages of observation methods and compare
them to survey methods;
6. Explain the concept of causality as defined in marketing research and
the concept of experimentation; and
7. Compare the use of laboratory versus field experimentation.

 INTRODUCTION
Analysis of secondary data helps define the marketing research problem and
develop an approach. Also, before the research design for collecting primary data
is formulated, the researcher should analyse the relevant secondary data. We will
discuss the distinction between primary and secondary data, the advantages and
disadvantages of secondary data and criteria for evaluating secondary data.
Internal secondary data is described, and major sources of external secondary
data such as published material, online and offline databases, will also be
discussed in this topic.
50  TOPIC 4 DATA COLLECTION METHOD AND SAMPLE DESIGN

Like secondary data analysis, qualitative research is a major methodology used

in exploratory research. Researchers undertake qualitative research to define
the problem or develop an approach. In developing an approach, qualitative
research is often used for generating hypotheses and identifying variables that
should be included in the research.

This topic will also discuss the types of survey methods that can be used in
research.

Lastly, this topic provides an overview of causal research and experimentation.

The concept and conditions for causality are described. Internal and external
validity in experimentation are also discussed in detail. A classification of
experimental designs will also be presented.

4.1 SECONDARY DATA

Secondary data is data that has already been collected for purposes other than
the problem at hand. Such data is often found internally but can also be found
externally in published materials, on computerised databases or from syndicated
services. Secondary data is characterised as being easily available and relatively
inexpensive.

4.1.1 Primary versus Secondary Data

Marketing data can be grouped into two types:
 Primary data; and
 Secondary data.

Primary data is information that is developed or gathered by the researcher,

particularly for the research project at hand. Secondary data has previously been
gathered by someone other than the researcher and/or for some other purpose
than the research project at hand. As government agencies and other organisations
gather information in the daily course of business, the gathered information that
is recorded is considered as secondary data. Table 4.1 shows the comparison
between primary data and secondary data.
 TOPIC 4 DATA COLLECTION METHOD AND SAMPLE DESIGN  51

Table 4.1: Comparison of Primary and Secondary Data

Aspect Primary Data Secondary Data

Collection purpose For the problem at hand For other problems
Collection process Very involved Rapid and easy
Collection cost High Relatively low
Collection time Long Short

Source: Malhotra, N. (2002). Basic marketing research: Applications to contemporary

issues. Upper Saddle River, New Jersey, USA: Prentice-Hall, Inc.

Businesses provide information to the government regarding sales by product or

service, number of employees, business location and so on. When a consumer
completes a warranty card to register a new washing machine, or television or
subscribes to a magazine, information is recorded. Much of this information is
valuable, and is provided to the public by government and private organisations.

The evolution of the Internet has done more to bring fast and easy access of
secondary data to end users than anything else. Since the mid-1980s, virtually
all documents have been electronically produced, edited, stored and made
accessible to users. For several years, firms have concentrated on bringing this
information to users through specialised services. Today, many of these firms
offer services via the Internet. Although some services are available only through
a subscription, the Internet provides an incredible stock of free secondary data.

EXERCISE 4.1

State the differences between primary data and secondary data.

4.1.2 Advantages and Disadvantages of Secondary

Data
There are four main advantages of using secondary data.

(a) First, it can be obtained quickly, in contrast to collecting primary data,

which may take several months from beginning to end.
52  TOPIC 4 DATA COLLECTION METHOD AND SAMPLE DESIGN

(b) Second, it is inexpensive as compared to collecting primary data.

(c) Third, it is usually available.

(d) Fourth, it enhances existing primary data.

Secondary data can help you:

(a) Identify the problem.

(b) Better define the problem.

(c) Develop an approach to the problem.

(d) Formulate an appropriate research design (for example, by identifying the

key variables).

(e) Answer certain research questions and test hypotheses.

(f) Interpret primary data more insightfully.

Although the advantages of secondary data almost always justify searching for
this information, there are caveats associated with secondary data. Some of the
problems associated with secondary data include:

(a) Lack of Availability

For some research questions, there is simply no available data. If Proton
wants to know the reaction of college students to a new two-seater sports
car design, it must show prototypes to the students and evaluate their
opinions. Of course, secondary data may have played a major role in the
engineerÊs design plan for the car.

(b) Mismatch of the Unit of Measurement

For example, Ahmad, a drugstore proprietor, finds that the primary
customers for his stores were families with a total annual household income
of between RM20,000 and RM40,000. Higher-income consumers tended to
purchase pricier drugs than Ahmad carried. When he was trying to decide
whether to open another store in Kuala Lumpur, he could not find useful
income data. One source offered class breakdowns in the following ranges:
RM30,000 to RM50,000; RM50,000 to RM70,000; RM70,000 to RM90,000 and
so forth.
 TOPIC 4 DATA COLLECTION METHOD AND SAMPLE DESIGN  53

(c) Inaccuracy
Before the secondary data can be used, the researcher should always assess
its accuracy. There are a number of potential sources of error when a
researcher gathers, codes, analyses and presents data. Any report that does
not mention possible sources and ranges of error should not be used.

(d) Insufficiency
A researcher may find that the available secondary data is relevant and
accurate but not sufficient to make a decision that will bring closure to
a problem. For example, a manager of Jaya Jusco may have sufficient
secondary data on income, family size, number of competitors and growth
potential to determine in which town in Selangor to locate the next Jaya
Jusco store. However, if no traffic counts exist for the selected towns,
primary data will have to be gathered to select a specific site for the store.

EXERCISE 4.2

Discuss the advantages and disadvantages of secondary data.

4.1.3 Criteria for Evaluating Secondary Data

Using secondary data does not relieve the researcher from attempting to assess
its accuracy. A few guidelines for determining the accuracy of secondary data are
as follows:

(a) Who Gathered the Data?

The source of the secondary data is the key to its accuracy. Federal
agencies, most state agencies and large commercial marketing research
firms generally can be counted on to have conducted their research as
professionally as possible.

(b) What was the Purpose or Objective of the Study?

Data is collected for certain reasons. Understanding the objectives of a
specific study can provide clues to the quality of the resulting data. There
have been situations in which advertising agencies have been hired by
clients to assess the impact of their own advertising programmes. In other
words, they have been asked to evaluate the quality of the job they are
doing for their clients.
54  TOPIC 4 DATA COLLECTION METHOD AND SAMPLE DESIGN

(c) What Information was Collected?

A good researcher must identify exactly what information was gathered
and from whom. For example, in a voter survey, were only „Barisan
Nasional‰ or „Pas‰ supporters interviewed? Were the respondents registered
voters? Was any attempt made to ascertain a respondentÊs likelihood of
voting in the next election? Was self-reported data used to infer actual
behaviour?

(d) When was the Information Collected?

A shopping mall study that surveyed shoppers only on weekends would
not reflect the opinions of „typical mall patrons.‰ A telephone survey
conducted from 9 am to 5 pm would vastly under represent working
persons.

(e) How was the Information Collected?

Was the data collected by mail, telephone, the Internet or personal interview?
Each of these techniques offers advantages and disadvantages.

(f) Is the Information Consistent with Other Information?

A researcher should be cautious of a lack of consistency between secondary
data sets and should investigate the possible cause of discrepancy.
Differences in the sample, time frame, sampling methodology, questionnaire
structure and other factors can lead to variations in the study.

4.1.4 Internal Secondary Data and Database

Marketing
Secondary data may be broadly classified as either internal or external (see
Figure 4.1). Internal secondary data is data that has been collected within the
firm. Such data includes sales records, purchase requisitions and invoices.
Obviously, a good marketing researcher always determines what internal
information is already available. We refer to internal data analysis as being part
of the internal report system of a firmÊs marketing information system (MIS).
 TOPIC 4 DATA COLLECTION METHOD AND SAMPLE DESIGN  55

Figure 4.1: Classifications of secondary data

Source: Malhotra, N. (2002). Basic marketing research: Applications to contemporary
issues. Upper Saddle River, New Jersey, USA: Prentice-Hall, Inc.

Today, a major source of internal data is databases that contain information on

customers, sales, suppliers and any other facets of business a firm may wish to
track. The term „database marketing‰ refers to the use of these internal databases
to target marketing programmes directly to consumers. The use of internal
databases has grown dramatically for several years.

Database marketing involves using computers to capture and track customer

profiles and purchase details. A database is a collection of data and information
describing items of interest. Each unit of information in a database is called a
record. A record could represent a customer, a supplier, a competitive firm, a
product, an individual inventory item and so on.

Internal databases are databases consisting of information gathered by a

company typically during the normal course of business transactions. Marketing
managers normally develop internal databases about customers but databases
may be kept on any topic of interest such as members of the sales force,
inventory, maintenance, and supplier firms. Companies use their internal
databases for purposes of direct marketing and to strengthen relationships with
customers through CRM or customer relationship management.
56  TOPIC 4 DATA COLLECTION METHOD AND SAMPLE DESIGN

4.1.5 Published Sources of External Secondary Data

External secondary data is data obtained from outside the firm. We can classify
the sources of external data into three groups:
(a) Published;
(b) Syndicated services ; and
(c) Databases.

Published sources are those sources of information that are prepared for public
distribution found in libraries or through a variety of other entities such as trade
associations, professional organisations, or companies. Published sources are
available in a number of formats including print, CD-ROM and online via the
Internet.

External databases are databases supplied by organisations outside the firm.

They may be used as sources of secondary data. Some of these databases are
available in printed format but, in recent years, many databases are now
available online. Online information databases are sources of secondary data
searchable by online search engines. Some online databases are available free of
charge and are supplied as a service by a host organisation. However, many
online information databases are available from commercial sources that provide
subscribers password (or IP address identification) access for a fee.

There are four types of databases categorised according to the nature of data
records within the databases:

(a) Bibliographic databases contain citations to journal articles, newspapers,

and government documents. Many bibliographic databases also include
full-text articles online. Examples include ABI Inform, an index covering
800 business, management and trade journals, and the Gale Academic
Index ASAP, an index which covers 960 multidisciplinary journals. Social
Science Citation Index, known as a citation bibliographic database, actually
indexes to bibliographies of articles allowing the users to learn which
articles „cited‰ certain sources.
 TOPIC 4 DATA COLLECTION METHOD AND SAMPLE DESIGN  57

(b) Numeric or statistical databases contain statistics or numeric data, such

as baseball game attendance records or sales for a given industry over a
specific time period. Examples include the 2010 Census of Population and
Housing and Country Business Patterns.

(c) Directory lists or databases list information about certain organisations,

individuals, government entities and so on. An example is GaleÊs Association
Unlimited, which lists more than 80,000 non-profit membership associations
in the United States.

(d) Comprehensive databases include many types of information. Examples

include Dow Jones Interactive, Gale Business and Company Resource Center
and LexisNexis.

4.2 QUALITATIVE RESEARCH

Qualitative research is a term used loosely to refer to research whose findings
are not subject to quantification or quantitative analysis. A quantitative study
may determine that heavy users of a particular brand of cigar range between
21 and 35 years of age and have an annual income of between RM24,000
and RM40,000. While quantitative research might be used to find statistically
significant differences between heavy users and light users, qualitative research
could be used to examine the attitudes, feelings and motivations of heavy users.

4.2.1 Primary Data: Qualitative versus Quantitative

Research
Table 4.2 compares qualitative and quantitative research at several levels.
Perhaps most significant to managers is the fact that qualitative research typically
is characterised by small samples ă a trait that has been a focal point for criticism
of all qualitative techniques. In essence, many managers are reluctant to base
important strategy decisions on small-sample research because it relies so greatly
on the subjectivity and interpretation of the researcher. They generally prefer a
large sample, with results analysed on a computer and summarised into tables.
These managers feel comfortable with marketing research based on large
samples and a high level of statistical significance because the data is generated
in a rigorous and scientific manner.
58  TOPIC 4 DATA COLLECTION METHOD AND SAMPLE DESIGN

Table 4.2: Qualitative versus Quantitative Research

Aspect Qualitative Research Quantitative Research

Types of questions Probing Limited probing
Sample size Small Large
Amount of information Substantial Varies
from each respondent
Hardware Interviewer with special Interviewer with fewer
skills special skills
Degree of replicability Subjective, interpretive Statistical, summation
Researcher training Low High
Scope of research Psychology, sociology, Statistics, decision modules,
social psychology, decision support system,
consumer behaviour, consumer programming,
marketing, marketing marketing, marketing
research research
Types of research Exploratory Descriptive or casual

Source: McDaniel, & Gates. (2002). Marketing research: The impact of the
internet (5th ed.), South-Western, Thomson Learning.

4.2.2 Classification of Qualitative Research Procedures

The procedures of qualitative research can be classified as either direct or
indirect, based on whether the respondents know the true purpose of the project.
In a direct approach, the purposes of the project are disclosed to the respondents
or they are obvious, given the nature of interview. On the other hand, in an
indirect approach, the purposes of the project are disguised from the
respondents.

Under a direct approach, the interviewer can use either focus group or in-depth
interview techniques.

(a) Focus Group Interview

A focus group interview is an interview with a small group of respondents
which consists of eight to 12 people, conducted by a trained moderator who
leads the discussion in a non-structured and natural manner. The main
purpose of a focus group is to gain insights on issues of interest to the
researcher by listening to a group of people from the appropriate target
market.
 TOPIC 4 DATA COLLECTION METHOD AND SAMPLE DESIGN  59

The following are basic steps for conducting a focus group:

(i) Designing the Environment

Focus group sessions are typically held in facilities specially equipped
to comfortably accommodate and record a group discussion. The
setting is typically an informal conference room equipped with a
one-way mirror and microphones throughout the room. Behind the
one-way mirror is a viewing room for management.

(ii) Recruiting and Selecting Focus Group Participants

A focus group is generally made up of eight to 12 members. A
focus group should be homogeneous in terms of demographic and
socioeconomic characteristics.

(iii) Selecting a Moderator

The moderator plays a key role and must be well trained. General
group management skills as well as a background in psychology and
marketing typically suffice in most focus group situations. The
moderator must be able to establish rapport with the participants.

(iv) Preparing the Discussion Guide

Regardless of the skills or qualifications of the moderator, any focus
group will flounder without an outline of the topics to be covered.
The discussion guide should reflect the objectives of the qualitative
research. A moderatorÊs discussion guide can reduce some of the
reliability problems inherent in focus groups such as the lack of
consistency in topics covered from group to group.

(v) Conducting the Group Interview

During the interview, the moderator must:

Ć Establish rapport with the group;

Ć State the rules of group interaction;

Ć Set objectives; and

Ć Attempt to summarise the groupÊs response to determine the

extent of agreement.

Ideally, the number of focus groups conducted should increase until

the moderator becomes familiar enough with the range of responses
and can anticipate what will be said.
60  TOPIC 4 DATA COLLECTION METHOD AND SAMPLE DESIGN

(vi) Preparing the Focus Group Report

Following the focus group session, the moderator and the managers
who viewed the session often engage in an instant interpretation.
There is value in this free exchange in that it captures the impressions
of the group and may be a good source of information for additional
brainstorming. Due to the small number of participants, frequencies
and percentages are not usually reported in a focus group summary.

Let us say that Tesco wanted to conduct a study on consumer

selection of hypermarkets. Beginning with the objectives of the
research project ă assessing the strengths and weaknesses of Tesco
hypermarket vis-à-vis competitors - the objectives of the qualitative
research are formed. Let us assume that the objective of the qualitative
research is to discover the primary factors that influence a consumerÊs
choice of a hypermarket. Based on this objective, several pertinent
questions can be asked including:

Ć How important is a hypermarket to consumers?

Ć How important is variety of selection?

Ć Are consumers willing to pay more if there is personal service? or

Ć How far are consumers willing to travel to shop?

Specific questions that are related to these general questions can then
be devised. For example, for the first question, the researcher can ask:

Ć Would you like to shop after 9pm? Why not?

Ć Do you like to have a salesperson help you select items when you
shop? or

Ć Do you often feel lost when looking for items? Should the store
help you in some way?

Based on the pertinent issues raised by the objectives of qualitative

research, a moderatorÊs outline can be devised.

Finally, although an actual focus group is probably impractical at this

stage, ideas for follow-up action can be generated. For example, based
on the focus group results, we can now proceed.
 TOPIC 4 DATA COLLECTION METHOD AND SAMPLE DESIGN  61

(b) In-Depth Interview Techniques

In-depth interview is an unstructured, direct and personal interview in
which a single respondent is probed by a highly skilled interviewer to
uncover underlying motivations, beliefs, attitudes and feelings on a topic.

In-depth interview can be effectively employed in special problem situations

such as those requiring:

(i) Detailed probing of the individuals;

(ii) Discussion on topics considered confidential, sensitive or embarrassing;

(iii) Situations where strong social norms exist and the individuals may be
easily swayed by group response;

(iv) Detailed understanding of complicated behaviour;

(v) Interview with professional people;

(vi) Interview with competitors who are unlikely to reveal information in

a group setting; or

(vii) Situations where the product consumption experience is sensory in

nature and affects moods, states and emotions.

Table 4.3 below shows the advantages and disadvantages of in-depth interviews.

Table 4.3: Advantages and Disadvantages of In-Depth Interviews

Advantages Disadvantages
 Great depth of insight can be  Skilled interviewers that are capable of
uncovered. conducting in-depth interviews are expensive
and difficult to find.
Ć The responses are associate Ć The lack of structure makes the results very
directly with the respondents. susceptible to the interviewerÊs influence.
Ć There is no social pressure to Ć The quality and completeness of the results
conform to group response. depend very heavily on the interviewerÊs skills.
ă Ć Data obtained is difficult to analyse and
interpret.

Under indirect approach, the interviewer can use projective techniques.

62  TOPIC 4 DATA COLLECTION METHOD AND SAMPLE DESIGN

Projective Techniques
Projective techniques use third-person association to indirectly ascertain the
respondentÊs motivations, attitudes and/or beliefs. When a personal and
threatening question is asked directly, it is assumed that the respondent will be
reluctant to answer truthfully, but if the same question is asked about a third
party, it is assumed that the respondent will answer more accurately and
completely. Projective techniques are used when it is believed that respondents
will not or cannot respond meaningfully to direct questions about (1) the reasons
for certain behaviour or attitudes, or (2) what the act of buying, owning or using
a product or service means to them. People may be unaware of their own feelings
and opinions, unwilling to make admissions that reflect badly on their self-image
(in which case, they will offer rationalisation or socially acceptable responses) or
too polite to be critical to an interviewer.

There are four forms of projective techniques with particular emphasis on word
association:

(a) Association Techniques

An individual is presented with a stimulus or stimuli and asked to respond
with the first thing or things that come to mind.

Word association is a projective technique in which respondents are

presented with a list of words, one at a time. After each word, they are
asked to give the first word that comes to mind. For example, the list
of items used as stimuli should include a random mix of neutral items such
as chair, sky and water, interspersed with the items of interest, such as
shopping downtown, vacationing in Malaysia or hamburger. An interviewer
reads the word to the respondents and asks them to mention the first thing
that comes to mind.

(b) Completion Techniques

The respondent is required to complete an incomplete stimulus situation.

(i) Sentence Completion

Sentence completion is a projective technique in which respondents
are presented with a number of incomplete sentences and asked to
complete them.

(ii) Story Completion

In this kind of projective technique, respondents are provided with
part of a story and required to give the conclusion in their own words.
 TOPIC 4 DATA COLLECTION METHOD AND SAMPLE DESIGN  63

(c) Construction Techniques

The respondent is required to construct a response in the form of story,
dialogue or description.

(i) Picture Response Technique

In this projective technique, the respondent is shown a picture and
asked to tell a story describing it.

(ii) Cartoon Test

This technique involves showing cartoon characters in a specific
situation related to the problem. The respondents are asked to
indicate the dialogue that one cartoon character might make in
response to the comment(s) of another character.

(d) Expressive Techniques

The respondents are presented with a verbal or visual situation and asked
to relate the feelings and attitudes of other people to the situation.

Role Playing
In this technique, respondents are asked to play a role or assume the
behaviour of someone else.

Projective techniques are used when the sensitivity of the subject matter is
such that respondents may not be willing or able to answer direct question
honestly. Projective techniques are also used to uncover subconscious
motives, beliefs or values, providing deeper insights and understanding as
part of exploratory research. Projective techniques are administered and
interpreted by trained interviewers who understand their advantages and
limitations.

EXERCISE 4.3

1. How would you find information about the 2010 Census?

2. Describe the purpose of the Population and Housing Census of
Malaysia 2010.
3. What are some of the disadvantages of data provided by public
data sources, such as sources published by the government?
64  TOPIC 4 DATA COLLECTION METHOD AND SAMPLE DESIGN

4.3 QUANTITATIVE RESEARCH

A survey research is a type of quantitative research method that utilises
a questionnaire to gather facts, opinions and attitudes of people called the
respondents. The information is gathered using either verbal or written questions.
Survey research collects information through mail, telephone calls or face-to-face
meetings. A sample survey emphasises that the purpose of contacting respondents
is to obtain a representative sample of the target population. Thus, a survey is
defined as a method of collecting primary data based on communication with a
representative sample of individuals. Figure 4.2 shows a classification of survey
methods.

Figure 4.2: A classification of survey methods

Source: Malhotra, N. (2004). Marketing research: An applied orientation.
Upper Saddle River, New Jersey, USA: Prentice-Hall, Inc.

4.3.1 Survey Methods Classified by Mode of

Administration
There are three major ways of collecting information from respondents:

(a) By having a person ask questions, either face-to-face or voice-to-voice

without any assistance from computers.

(b) By having computer assistance or direct questioning in a face-to-face or

voice-to-voice survey.

(c) By allowing the respondents to fill out the questionnaires themselves,

without computer or personal assistance.
 TOPIC 4 DATA COLLECTION METHOD AND SAMPLE DESIGN  65

These three alternatives are referred to as:

(a) Person-administered survey;
(b) Computer-administered survey; and
(c) Self-administered survey.

Each method has its special advantages and disadvantages.

4.3.2 A Person-administered Survey

A person-administered survey is a method of data collection in which the
interviewer reads the questions to respondents either face-to-face or over the
telephone and records his or her answers.

(a) Telephone Interviews

This survey method involves telephoning a sample of respondents and
asking them a series of questions.

(b) Door-to-door Interviews

This type of interview refers to personal interviews which may be
conducted at respondentsÊ homes or offices, or in many other places.

(c) Mall-Intercept Personal Interviews

Mall-intercept interviews allow many interviews to be conducted quickly.
Respondents are intercepted while they are shopping in a mall and brought
to a central location test facility in the mall. The interviewer administers the
questionnaire in a similar manner to that of the in-home personal survey.

4.3.3 A Computer-administered Survey

Computer technology and the continuing developments in this field offer viable
options with respect to survey mode. Although person-administered surveys are
still the industry mainstay, computer-administered methods are growing very
rapidly and will rival person-administered surveys in the foreseeable future. A
computer-administered survey is one in which computer technology plays an
essential role in the interview work.

(a) Computer-assisted Personal Interviews (CAPI)

The respondents sit in front of a computer terminal and answer the
questionnaire on the screen by using the keyboard and/or mouse.
66  TOPIC 4 DATA COLLECTION METHOD AND SAMPLE DESIGN

(b) Computer-assisted Telephone Interviews (CATI)

The computer dials a telephone number, the interviewer asks the questions
on the screen of the computer, and the interviewer records the respondentsÊ
answers directly.

(c) Electronic Interviewing

Electronic interviewing can be conducted either by e-mail or administered
on the Internet:

(i) E-mail Interview

The survey is written within the body of the e-mail message, which is
then sent out over the Internet. E-mail surveys use pure text (ASCII)
to represent questionnaires and can be received and responded to by
anyone with an e-mail address, whether or not they have access to the
Internet.

(ii) Internet Interview

Internet and web surveys use hypertext markup language (HTML),
the language of the web, and are posted on a website. Respondents
can be recruited over the Internet from potential respondentsÊ
databases maintained by the marketing research firm or they can be
recruited through conventional methods (mail, telephone).

4.3.4 A Self-administered Survey

A self-administered survey is one in which the respondents complete the surveys
themselves. It is different from other survey methods as there is no agent ă
human or computer ă administering the interview.

(a) Door-to-door Self-Administered

Self-administered interviews may be conducted at the respondentsÊ home
or offices, or in other places.

(b) Mall Intercept

Self-administered surveys can be conducted at shopping malls, where
the respondents are intercepted and are asked to complete the survey
questionnaires without any assistance from the researcher.
 TOPIC 4 DATA COLLECTION METHOD AND SAMPLE DESIGN  67

(c) Mail Interview

Questionnaires are mailed to pre-selected potential respondents, who then
complete and return the questionnaires by mail.

(d) Mail Panel

Consists of a large and nationally representative sample of households who
have agreed to periodically participate in mail questionnaires, product tests
and telephone surveys conducted by the sponsoring organisations.

4.3.5 Criteria for Selecting Survey Methods

Table 4.4 explains the criteria for selecting survey methods.

Table 4.4: Criteria for Selecting Survey Methods

Criteria Explanations
(a) Flexibility of data The flexibility of data collection is determined primarily by
collected the extent to which the respondents can interact with the
interviewer and the survey questionnaire. The personal
interview, whether conducted in homes or as a mall-intercept
interview, allows the highest flexibility of data collection. CATI
or CAPI and Internet surveys allow somewhat greater
flexibility because the questionnaire is administered in an
interactive mode. Since they do not allow for interaction
between the interviewer and the respondent, mail surveys, mail
panels and e-mail surveys have low flexibility.
(b) Diversity of The diversity of questions that can be asked in a survey
questions depends on the degree of interaction the respondent has with
the interviewer and the questionnaire, as well as the ability to
actually see the questions. Therefore, in-home, mall intercept
and CAPI allow for diversity. In Internet surveys, multimedia
capabilities can be utilised so that the ability to ask diverse
questions is moderate to high, despite the absence of an
interviewer. In mail surveys, mail panels and e-mail surveys,
less diversity is possible. In traditional telephone interviews
and CATI, the respondents cannot see the questions while
answering, and this limits the diversity of the questions.
68  TOPIC 4 DATA COLLECTION METHOD AND SAMPLE DESIGN

(c) Use of physical Often it is helpful or necessary to use physical stimuli such as
stimuli the product, a product prototype, commercials or promotional
displays during the interview. In these cases, personal interviews
conducted at central locations (mall intercept and CAPI) are
preferable to in-home interviews. Mail surveys and mail panels
are moderate on this dimension, because it is sometimes
possible to mail the facilitating aids or even product samples.
Internet surveys are also moderately suitable. Since they are
web-based, the questionnaires can include multimedia elements
such as prototype web pages and advertisements. The use of
physical stimuli is limited in traditional telephone interviews,
CATI and e-mail surveys.
(d) Quantity of data In-home personal interviews allow the researcher to collect
large amounts of data. In contrast to in-home interviews, mall
intercepts and CAPI provide only moderate amount of data.
Mail surveys also yield moderate amount of data. The same is
true for e-mail and Internet surveys, although the Internet is a
better medium in this respect. Mail panels, on the other hand,
can generate a large amount of data because of the special
relationships between the panel members and the sponsoring
organisations. Traditional telephone interviews and CATI
results in the most limited quantities of data.
(e) Sample control Sample control is the ability of the survey mode to reach the
units specified in the sample effectively and efficiently. In-home
personal interviews offer the best sample control. Mall-intercept
interviews allow only a moderate degree of sample control.
Compared to a mall intercept, CAPI offers slightly better
control, as sampling quotas can be set and respondents
randomised automatically. Moderate to high sampling control
can be achieved with traditional telephone interviews and
CATI. The degree of sample control in mail surveys is low. Mail
panels, on the other hand, provide a moderate to high degree of
control over the sample. They provide samples matched to the
Malaysian population. However, not all the population are
candidates for Internet survey research. Thus, sample control is
low to moderate for Internet surveys. E-mail surveys suffer
from many of the limitations of mail surveys and, thus, offer
low sample control.
(f) Control of data Personal interviews conducted at central locations (mall
collection intercept and CAPI) offer the greatest degree of environmental
environment control. In-home personal interviews, traditional telephone
and CATI offer moderate control. In mail surveys and panels,
e-mail and Internet, the researcher has little control over the
environment.
 TOPIC 4 DATA COLLECTION METHOD AND SAMPLE DESIGN  69

(g) Control of field The field force consists of interviewers and supervisors involved
force in data collection. Since they require no such personnel, mail
survey, mail panels, e-mail surveys and Internet surveys
eliminate field force problems. Traditional telephone interviews,
CATI, mall intercept and CAPI all offer a moderate degree of
control because the interviewers are problematic in this respect.
(h) Response rate Personal, in-home, mall intercept and computer-assisted
interviews yield the highest response rate (typically more than
80 percent). Telephone interviews, traditional and CATI, yield
response rates between 60 and 80 percent.
Mail surveys have poor response rates. However, response
rates in mail panels are typically in the 70 to 80 percent range
because of assured respondentsÊ cooperation. Internet surveys
have the poorest response rate, even lower than e-mail surveys.
This is due to the fact that some respondents may have access
to e-mail but not the Internet and assessing the Internet requires
more effort and skills.
(i) Perceived This refers to respondentsÊ perceptions that the interviewer
anonymity or the researcher will not discern their identity. Perceived
anonymity of the respondents is high in mail surveys, mail
panels and Internet surveys because there is no contact with the
interviewer while responding. It is low in personal interviews
(in-home, mall intercept and computer assisted) due to face-
to-face contact with the interviewer. Traditional telephone
interviews and CATI fall in the middle. It is also moderate
with e-mail because although there is no contact with the
interviewer, respondents know that their names can be located
on the return e-mail.
(j) Social As in mail surveys, mail panels and internet surveys do not
desirability/ involve any social interaction between the interviewer and the
sensitive respondent, they are the least susceptible to social desirability.
information Traditional telephone interviews and CATI are moderately
good at avoiding socially desirable responses. They are good
for obtaining sensitive information, as the respondents perceive
that they are not committing to anything in writing over the
telephone. E-mail is only moderately good for controlling social
desirability and obtaining sensitive information, given the
respondentsÊ awareness that their names can be located in
the return e-mail. Personal interviews, whether in-home, mall
intercept or computer assisted, are limited in this respect,
although the problem is somewhat reduced in the case of
computer-assisted interviews.
70  TOPIC 4 DATA COLLECTION METHOD AND SAMPLE DESIGN

(k) Potential for The extent of the interviewerÊs role determines the potential
interviewer bias for bias. In-home and mall-intercept personal interviews are
highly susceptible to interviewer bias. Traditional telephone
interviews and CATI are less susceptible, although the potential
is still there. Computer-assisted personal interviews have a low
potential for interviewer bias. Mail surveys, mail panels, e-mail
and Internet surveys are free of interviewer bias.
(l) Speed The Internet is by far the fastest method, followed by e-mail.
Traditional telephone interviews and CATI are also fast ways of
obtaining information. Next in speed are mall-intercept and
computer-assisted personal interviews, which reach potential
respondents in central locations. In-home personal interviews
are slower, because there is down-time between interviews
while the interviewer travels to the next respondentÊs home.
Mail surveys are typically the slowest. Mail panels are faster
than mail surveys, as little follow-up is required.
(m) Cost In general, Internet, e-mail, mail surveys, mail panels,
traditional telephone, CATI, CAPI, mall intercept and personal
in-home interviews require progressively larger field staff and
greater supervision and control. Hence, the cost increases in the
order given.

4.3.6 Improving Survey Response Rates

What are the methods that can be used to improve the survey response rate?
Regardless of the survey methods chosen, response rates can be improved by:

(a) Prior Notification

Notification involves sending a letter or e-mail, or making a telephone call
to potential respondents, notifying them of the imminent mail, telephone,
personal or electronic survey.

(b) Incentives
Offering monetary as well as non-monetary incentives to potential
respondents can increase response rates. Monetary incentives can be
prepaid or promised.

(c) Follow-up
Follow-up, or contacting the non-respondents periodically after the initial
contact, is particularly effective in decreasing refusal in mail survey.
 TOPIC 4 DATA COLLECTION METHOD AND SAMPLE DESIGN  71

(d) Other Facilitators of Response

Personalisation, or sending letters addressed to specific individuals, is
effective in increasing response rates.

Figure 4.3 illustrates the methods of improving response rates.

Figure 4.3: Methods of improving response rates

Source: Malhotra, N. (2002). Basic marketing research: Application to
contemporary issues. Upper Saddle River, New Jersey, USA: Prentice-Hall, Inc.

4.4 OBSERVATION METHODS

Observation methods can be classified as personal observation and mechanical
observation (see Figure 4.4). Personal observation is the situation where the
researcher personally observes actual behaviour as it occurs, and the observer
merely records what takes place. The researcher does not attempt to control or
manipulate the phenomenon being observed. On the other hand, mechanical
observation is the condition where devices are used to continually record on-
going behaviour for later analysis.
72  TOPIC 4 DATA COLLECTION METHOD AND SAMPLE DESIGN

Figure 4.4: A classification of observation methods

Source: Malhotra, N. (2002). Basic marketing research: Application
to contemporary issues. Upper Saddle River, New Jersey, USA: Prentice-Hall, Inc.

(a) Structured observation is appropriate when the marketing problem

has been clearly defined and the information needed has been specified. It
requires the researcher to specify in detail what is to be observed and
how the measurements are to be recorded. Unstructured observation is
appropriate when the problem has yet to be formulated precisely and
flexibility in observation is needed to identify key components of the
problem and develop hypotheses.

(b) Disguised observation is used when the respondents might act differently
under direct observation, for example, the purchase of sensitive or personal
products. In disguised observation, the respondents are not aware that they
are being observed, whereas in undisguised observation, the respondents
are aware that they are under observation.

(c) Natural observation involves observing behaviour as it normally takes

place in the environment. Contrived observation takes place in an artificial
environment that has been created by the researcher. Natural observation
allows the researcher to observe actual behaviour, but does not allow for
inferences on the causes of behaviour. Contrived observation enables the
researcher to control more of the variables that affect behaviour.

4.4.1 Relative Advantages and Disadvantages of

Observation Methods versus Survey Methods
Table 4.5 shows the advantages and disadvantages of observation methods
versus survey methods.
 TOPIC 4 DATA COLLECTION METHOD AND SAMPLE DESIGN  73

Table 4.5: Advantages and Disadvantages of Observation Methods

versus Survey Methods

Advantages Disadvantages
Observational methods permit measurement Very little can be inferred about the
of actual behaviour. motives, beliefs, attitudes and preferences
underlying the observed behaviour.
The potential bias caused by the interviewer The observerÊs selective perceptions can
and the interviewing process is eliminated bias the data.
or reduced.
Certain types of data can be collected best In some cases, the use of observational
only by observation. methods may border on being unethical
because subjectsÊ behaviour is being
monitored without their explicit knowledge
or consent.
If the observed phenomenon occurs at ă
relatively frequent intervals and is of short
duration, observational methods may cost
less and be faster than survey methods.

Source: Malhotra, N. (2002). Basic marketing research: Application to

contemporary issues. Upper Saddle River, New Jersey, USA: Prentice-Hall, Inc.

4.4.2 Comparison of Survey and Observation Methods

What do you think are the differences between survey and observation? There
are several advantages to using observational data collection methods. First, they
do not require conscious respondentsÊ participation, which minimises non-
response errors. Although ethical questions surround the practice of observation
without consent, even conscious participation requires less effort from the
respondents than that required with other research techniques.

Interviewer bias resulting from interaction with the respondents or subjective

interpretation of the questionnaires is minimised because the observer only has
to record what is occurring. Additionally, the errors inherent in self-reported
behaviour are eliminated given that the observer records only actual behaviour
without asking the respondents any questions.
74  TOPIC 4 DATA COLLECTION METHOD AND SAMPLE DESIGN

Observational data provides insight into what behaviour is occurring, but not
why. Attitudes, motivations and values are all lost to the observational method.
Additionally, highly personal behaviours related to personal hygiene or family
interactions are not available for observation.

Individuals have a tendency to only observe what they want to, and this may
cause an observer to overlook important aspects of behaviour. This perceptual
difference among observers threatens the integrity of the approach. Finally,
observational techniques can be adopted for only frequent behaviours of short
duration. Behaviours occurring infrequently or spanning a long period of time
are too expensive to record using this technique.

To sum up, observation can potentially provide valuable information when

properly used. From a practical standpoint, it is best to view observation as a
complement to survey methods, rather than as being in competition with them.

EXERCISE 4.4

1. List the major advantages of survey research methods over

qualitative methods. Identify the major drawbacks.

2. In a short essay, discuss the characteristics of the survey method of

obtaining information, and list at least three advantages and three
disadvantages of survey research.

3. List and briefly discuss the three categories of personal

interviewing methods.

4. List and briefly discuss the three primary methods researchers use
in an attempt to improve response rates.

5. List and briefly discuss the two major observation methods used
in descriptive research. Next, list at least one advantage and one
disadvantage of each of the two major observation methods.
 TOPIC 4 DATA COLLECTION METHOD AND SAMPLE DESIGN  75

4.5 CONCEPT OF CAUSALITY

Experimentation is commonly used to infer causal relationship. The concept
of causality requires some explanation. The scientific concept of causality is
complex. Causality means something very different to the average person on the
street than it does to a scientist.

The scientific meaning of causality is more appropriate to marketing research

than is the everyday meaning. Marketing effects are caused by multiple
variables, and the relationship between cause and effect tends to be probabilistic.
Moreover, we can never prove causality, we can only infer a cause-and-effect
relationship.

4.5.1 Conditions for Causality

There are three necessary conditions to infer causality:

(a) Concomitant Variation

It is the extent to which a cause, X, and an effect, Y, occur together or
vary together in the way predicted by the hypothesis under consideration.
Evidence of concomitant variation can be obtained in either a qualitative or
quantitative manner. Thus, we may hypothesise that increasing the product
price will lead to an increase in profits.

(b) Time Order of Occurrence of Variables

It states that the causing event must occur either before or simultaneously
with the effect; it cannot occur afterwards. Thus, profits cannot increase
until the price increases.

(c) Absence of Other Possible Factors

It means that we can only infer causality if all other factors affecting the
subject are absent (i.e., these factors are held constant). Thus, other factors
such as store penetration, expanded consumer spending or decreased cost
must be held constant.

EXERCISE 4.5

What are the conditions for causality?

76  TOPIC 4 DATA COLLECTION METHOD AND SAMPLE DESIGN

4.6 WHAT IS EXPERIMENTATION?

Experimentation is the process of manipulating one or more independent
variables and measuring their effect on one or more dependent variables, while
controlling for the extraneous variables.

(a) Independent Variables

Variables or alternatives that are manipulated and whose effects are
measured and compared, such as price levels, different package designs
and advertising themes.

(b) Test Units

Individuals, organisations or other entities whose response to the dependent
variables or treatments is being examined such as consumers, store size,
store location and competitive effort.

(c) Dependent Variables

These are the variables that measure the effect of the independent variables on
the test units. These variables may include sales, profits and market shares.

(d) Extraneous Variables

These are all variables other than the independent variables that affect
the response of the test units. These variables can confound the dependent
variable measures in a way that weakens or invalidates the results of
experiment. Extraneous variables include store size, store location and
competitive effort.

(e) Experiment
An experiment is formed when the researcher manipulates one or more
independent variables and measures their effect on one or more dependent
variables, while controlling the effect of extraneous variables.

(f) Experimental Design

It is a set of procedures specifying:

(i) The test units and how these units are to be divided into homogenous
sub-samples;

(ii) Independent variables or treatments that are to be manipulated; and

(iii) How to deal with the extraneous variables.

 TOPIC 4 DATA COLLECTION METHOD AND SAMPLE DESIGN  77

4.6.1 Validity of Experimentation

What are the variables that can affect validity?

In an experimental setting, validity is the ability to draw appropriate conclusions

about the effects of independent variables and to make reasonable generalisations
to a larger population of interest. To continue this definition, internal validity
examines whether the manipulation of the independent variables or treatments
actually caused the effects on the dependent variables, whereas external validity
examines whether the cause-and-effect relationship found in the experiment can
be generalised.

Since it is difficult to understand the concept of validity, some examples of a

research design lacking one of the two types of validity would be helpful. For
instance, if we hypothesise that selling through a national chain will increase
profit, we must hold all other causes of increased profit constant to conclude
internal validity. External validity can be seen in a study that randomly selects
stores from all markets in which a national chain operates. Assuming that the
study takes into account all realistic factors, any finding that holds for all these
markets can be generalised to all stores in the national chain.

4.6.2 A Classification of Experimental Designs

Figure 4.5 below shows that experimental designs are divided into four
categories of design.

Figure 4.5: Classification of experimental designs

Source: Malhotra, N. (2004). Marketing Research: An Applied Orientation. Upper Saddle
River, New Jersey, USA: Prentice-Hall, Inc.
78  TOPIC 4 DATA COLLECTION METHOD AND SAMPLE DESIGN

(a) Pre-experimental Designs

This type of design does not apply randomisation; thus, the results can be
influenced by extraneous factors.

(b) True Experimental Designs

Randomisation is included in these designs; thus, researchers are allowed
to randomly assign test units and treatments to experimental groups.

(c) Quasi-experimental Designs

This kind of design lacks control over the scheduling of treatment and the
ability to randomly expose test unit to treatments.

(d) Statistical Designs

These involve a series of basic experiments that are conducted
simultaneously. These designs suffer from the same shortcomings as the
actual designs of the basic experiments.

4.6.3 Laboratory versus Field Experiments

Experiment may be conducted in laboratory or field environments. A laboratory
environment is an artificial one, which the researcher constructs with the desired
conditions specific to the experiment. The term field experiment is synonymous
with actual market conditions.

Laboratory experiments have some advantages over field experiments as listed

below.

(a) The laboratory environment offers a high degree of control because it

isolates the experiment in a carefully-monitored environment. Therefore,
the effects of history can be minimised.

(b) A laboratory experiment also tends to produce the same results if repeated
with similar subjects, leading to high internal validity.

(c) Laboratory experiments tend to use small numbers of test units, last for a
shorter time, be more restricted geographically and be easier to conduct
than field experiments. Hence, they are generally less expensive as well.
 TOPIC 4 DATA COLLECTION METHOD AND SAMPLE DESIGN  79

As compared to field experiments, laboratory experiments suffer from some

disadvantages listed below.

(a) The artificiality of the environment may cause reaction error, in that the
respondents react to the situation itself, rather than to the independent
variable.

(b) The environment may cause demand artifacts, a phenomenon in which the
respondents attempt to guess the purpose of the experiment and respond
accordingly.

(c) Laboratory experiments are more likely to have lower external validity
than field experiments. Since a laboratory experiment is conducted in an
artificial environment, the ability to generalise the results to the real world
may be diminished.

The researcher must consider all of these factors when deciding whether to
conduct laboratory or field experiments. Field experiments are less common in
marketing research than laboratory experiments, although laboratory and field
experiments play complementary roles.

4.6.4 Limitations of Experimentation

Experimentation is a very powerful way to uncover causal relationship.
However, in practice, real world considerations prevent it from achieving its
optimal effectiveness. These considerations are:

(a) Time
Experiment can be time consuming; however, the experiments should be
long enough in duration so that the post-treatment measurement includes
most, or all, effects of the variables.

(b) Cost
The requirements of experimental group, control group and multiple
measurements significantly increase the cost of the experiment.

(c) Administration
Experiment can be difficult to administer, and it is not always feasible to
control for the effects of the extraneous variables.
80  TOPIC 4 DATA COLLECTION METHOD AND SAMPLE DESIGN

EXERCISE 4.6

Briefly describe the four broad categories of experimental designs and

discuss the various applications of each design category.

 In contrast to primary data that is collected for a specific purpose, secondary

data has been generated for reasons other than the current research problem.
The cost advantage and speed with which secondary data can be obtained
must be weighed against its limited fit to the current research problem and
concerns regarding data accuracy.

 Secondary data can be found both inside and outside the firm. Technological
advances are facilitating the establishment of sophisticated internal databases
and have made external secondary data just a mouse click away.

 Qualitative research is appropriate when a company is faced with an

uncertain situation or is presented with conclusive results that differ from
expectations. This type of research method can provide insight either before
or after the fact.

 Survey and observation are the two primary methods of conducting

quantitative descriptive research.

 Survey involves the direct questioning of respondents, while observation

involves recording respondentsÊ observed behaviour.

 Survey involves the administration of questionnaires and may be classified,

based on the methods or mode of administration, as traditional telephone
interviews, computer-assisted telephone interviews (CATI), in-home personal
interviews, mall-intercept interviews, computer-assisted personal interviews
(CAPI), mail survey, mail panels, and electronic surveys administered by
e-mail or Internet.
 TOPIC 4 DATA COLLECTION METHOD AND SAMPLE DESIGN  81

 The major observational methods are personal observation and mechanical

observation. As compared to surveys, the relative advantages of observational
methods are:
ă They permit measurement of actual behaviour;
ă There is no reporting bias; and
ă There is less potential for interviewer bias.

 Certain types of data can be obtained best only by observation. The relative
disadvantages of observation are:
ă Very little can be inferred about motives, beliefs, attitudes, and preferences;
ă There is potential for observer bias;
ă Most methods are time consuming and expensive;
ă It is difficult to observe some forms of behaviour; and
ă There is potential for unethical behaviour.

 The scientific notion of causality implies that we can never prove that X
causes Y. At best, we can only infer that X is one of the causes of Y in that it
makes the occurrence of Y probable.

 Three conditions must be satisfied before causal inferences can be made:

ă Concomitant variation, which implies that X and Y must vary together in
the hypothesised way;
ă Time order of occurrence of variables, which implies that X must precede
Y; and
ă Elimination of other possible causal factors, which implies that competing
explanations must be ruled out.

 Experiments provide the most convincing evidence for all three conditions.
An experiment is formed when one or more independent variables are
manipulated or controlled by the researcher, and their effects on one or more
dependent variables are measured.
82  TOPIC 4 DATA COLLECTION METHOD AND SAMPLE DESIGN

 In designing an experiment, it is important to consider internal and

external validity. Internal validity refers to whether the manipulation of the
independent variables actually caused the effects on the dependent variables.
External validity refers to the generalisability of experimental results.

 Experimental design may be classified as pre-experimental, true experimental,

quasi-experimental or statistical. An experiment may be conducted in a
laboratory environment or under actual market conditions in a real-life
setting. Only causal designs encompassing experimentation are appropriate
for inferring cause-and-effect relationships.

A computer-administered survey Projective techniques

A person-administered survey Primary data
A self-administered survey Response rate
Consumer diary panels Secondary data
Database marketing Survey
In-depth interview Syndicate data
Observation method
Topic  Measurement
and Scaling
5
LEARNING OUTCOMES
By the end of this topic, you should be able to:
1. Describe the concepts of measurements of scaling;
2. Compare the primary scales of measurement and differentiate
nominal, ordinal, interval and ratio scales;
3. Analyse the scaling techniques as comparative and non-comparative;
and
4. Examine each type of comparative scaling and non-comparative
scaling technique.

 INTRODUCTION
You are having breakfast at your favourite restaurant, and a researcher asks you
to taste the new nasi lemak dish. You sample it, and the researcher asks you
to complete a short survey. You are to rate the nasi lemak using the following
scale: (5) strongly agree, (4) agree, (3) neither agree not disagree, (2) disagree, or
(1) strongly disagree in response to statements on the survey. Here are some
sample statements:

(a) The nasi lemak is good.

(b) The nasi lemak sambal is just right for me.

(c) I intend to order this nasi lemak the next time I visit this restaurant.
84  TOPIC 5 MEASUREMENT AND SCALING

In the example, although a scale of „strongly agree‰ to „strongly disagree‰ is

used to rate the nasi lemak, the responses can be effectively analysed only by
assigning the number to them. These numbers aid in measuring the responses.
Measurement means assigning numbers to characteristics according to specified
rules to reflect the quantity of the characteristics that the test products possess. In
our nasi lemak survey, a „5‰ represents strongly agree, a „4‰ represents agree
and so on.

Therefore, this topic will describe the concepts of scaling and measurement. In
this topic, we will distinguish the four primary scales of measurement:
(a) Nominal;
(b) Ordinal;
(c) Interval; and
(d) Ratio.

We will next discuss comparative and non-comparative scaling techniques.

Each type of comparative scaling technique: ranking from paired comparison,
rank order and constant sum, will be discussed in detail. The types of non-
comparative scaling techniques such as continuous and itemised rating scales
will also be discussed.

The topic will also discuss briefly non-comparative itemised rating scale
decisions that are available to the researcher.

5.1 MEASUREMENT AND SCALING

Measurement precedes scaling in test construction. Measurement is the
assignment of numbers or other symbols to the characteristics of objects
according to certain pre-specified rules. Scaling is an extension of measurement
that involves the generation of a continuum on which measured objects are
located.
 TOPIC 5 MEASUREMENT AND SCALING  85

5.1.1 Primary Scales and Level of Measurement

Now, let us look at primary scales of measurement. Figure 5.1 shows the four
types of primary scales of measurement.

Figure 5.1: Primary scales of measurement

(a) Nominal Scale

This is used only as labelling scheme in which numbers serve only as
labels or tags for identifying and classifying objects (objects are mutually
exclusive and exhaustive). The numbers in a nominal scale do not reflect
the amount of a characteristic that the objects possess; rather, they are used
only for identification. Some examples are numbers on football playersÊ
uniforms, street names or identification card numbers.

Example:
1. Your gender
_______________ Male
_______________ Female

2. Your department
_______________ Production
_______________ Sales
_______________ Accounting
_______________ Finance
_______________ R&D
_______________ Other (specify)
86  TOPIC 5 MEASUREMENT AND SCALING

(b) Ordinal Scale

This is a ranking scale in which assigned numbers indicate the relative
extent to which an object possesses some characteristics. It can be
determined whether that object has more or less of the characteristics than
other objects. Some examples are the ranking of teams for the Malaysian
Golf tournament, socioeconomic status and quality rankings.

Example 1:
Rank the following personal computers with respect to their usage in your
office, assigning the number 1 to the most used system, 2 to the next most used
and so on. If a particular system is not in your office, put 0 next to it.

Apple Hewlett Packard

Compaq IBM
Acer Packard Bell
Dell Computer Sony
Gateway Toshiba
Other (specify)

Example 2:
Rank the states below in the order that you consider them suitable as a
location for a new plant. The state considered the most suitable should be
ranked 1, the next ranked 2 and so on.

Johor Selangor
Perak Pahang
Melaka Terengganu
Kedah

(c) Interval Scale

Numbers are used to rate objects such that numerically equal distances
on the scale represent equal distances in the characteristics being measured.
Examples include time and temperature.
 TOPIC 5 MEASUREMENT AND SCALING  87

Example:
Using the scale below, please indicate your response to each of the items
that follow by circling the number that describes your feeling.

Neither
Strongly Strongly
Disagree Agree nor Agree
Disagree Agree
Disagree
1 2 3 4 5

(a) My job offers me a chance to test myself and 1 2 3 4 5

my abilities.
(b) Mastering this job would mean a lot to me. 1 2 3 4 5
(c) Doing this job well is a reward in itself. 1 2 3 4 5
(d) Considering the time spent on the job, I feel 1 2 3 4 5
thoroughly familiar with my task and
responsibilities.

(d) Ratio Scale

This scale is used to identify objects, rank order to objects and compare
intervals or differences, for such characteristics as height, age and income.

Example:

1. How many other organisations did you work for before joining the
system?

2. Please indicate the number of children you have in each of the following
categories:
_______________ below 3 years of age
_______________ between 3 and 6 years of age
_______________ between 6 and 12 years of age
_______________ 12 years and above

3. How many retail outlets do you operate?

88  TOPIC 5 MEASUREMENT AND SCALING

Table 5.1 shows some commonly used examples of primary scale.

Table 5.1: Primary Scales of Measurement

Common Marketing Permissible

Scale Basic Characteristics
Examples Examples Statistics
Nominal Numbers identify Identification Brand Percentages,
and classify objects. Card number, numbers, store mode.
numbering of types, sex
football classification.
players.
Ordinal Numbers indicate the Quality Preference Percentile,
relative positions of rankings, ranking, median.
the objects but not the ranking of market
magnitude of teams in a position, social
difference between tournament. class.
them.
Interval Differences between Temperature Attitudes, Range, mean,
objects can be (Fahrenheit, opinions, index standard
compared; zero point Celsius). numbers. deviation.
is arbitrary.
Ratio Zero point is fixed; Length, weight. Age, income, Geometric
ratios of scale values costs, sales, mean (All).
can be computed. market shares.

Source: Malhotra, N. (2004) Marketing research: An applied orientation.

Upper-Saddle River, New Jersey, USA: Pearson Education Inc.

Figure 5.2 shows a classification of scaling techniques.

Figure 5.2: A classification of scaling techniques

Source: Malhotra, N. (2004) Marketing research: An applied orientation.
Upper-Saddle River, New Jersey, USA: Pearson Education Inc.
 TOPIC 5 MEASUREMENT AND SCALING  89

5.2 COMPARATIVE SCALING TECHNIQUES

Comparative scaling techniques involve a direct comparison of two or more
objects. For example, respondents might be asked whether they prefer Coke
or Pepsi. A comparative scale gives the marketer data that measures relative
differences. It has only ordinal or rank-order properties.

More Useful About the same Less Useful

1 2 3 4 5

5.2.1 Paired Comparison Scaling

In paired comparison scaling, a respondent is presented with a pair of
alternatives and asked to select one according to some criterion. Data obtained in
this way is ordinal in nature. For example, a consumer involved in a paired
comparison study may state that she shops in Tesco more than in Giant, or that
she likes Nokia mobile phones more than Samsung mobile phones.

Paired comparison scaling is useful when the number of brands under

consideration is limited to no more than five. When a large number of brands are
involved, the number of comparisons becomes bulky.

Example:
I would like to know your overall opinion of two brands of adhesive bandages.
They are Curad and Band-Aid. Overall, which of these two brands ă Curad or
Band-Aid ă do you think is the better one?
Curad is better ________________
Band-Aid is better ________________
They are the same ________________

5.2.2 Rank Order Scaling

In rank order scaling, the respondents are presented with several alternatives
simultaneously and are asked to rank them according to some criterion. For
example, consumers may be asked to rank brands of jeans according to overall
preferences. These rankings are typically obtained by asking the respondents to
assign a rank of 1 to the most preferred brand, 2 to the second most preferred and
so on, until each alternative is ranked down to the least preferred brand.
90  TOPIC 5 MEASUREMENT AND SCALING

Rank order scaling is commonly used to measure preferences among brands

as well as among brand attributes. Rank order scaling forces the respondents
to discriminate along alternatives. This type of scaling process comes closer to
resembling the shopping environment.

Example:
Rank the various brands of personal computer in order of preference. Begin by
picking out the one brand you like most and assign it a number 1. Then find the
second most preferred brand and assign it a number 2. Continue this procedure
until you have ranked all the brands of personal computers in order of
preference. The least preferred brands should be assigned a rank of 10. No brand
should receive the same rank number. The criterion of preference is entirely up to
you. There is no right or wrong answer. Just try to be consistent.

Brand Rank Order

Acer
Dell
Apple
IBM
Compaq
Hewlett Packard
Gateway
Packard Bell
Sony
Toshiba

5.2.3 Constant Sum Scaling

In constant sum scaling, respondents allocate a constant sum of units, such as
points, dollars or chips, among a set of alternatives according to some specified
criterion. Respondents may be asked, for example, to allocate 100 points to eight
attributes of a bath soap. The points are allocated to represent the importance
attached to each attribute. If an attribute is unimportant, the respondent assigns
it zero points; if an attribute is twice as important as some other attributes, the
respondents assign it twice as many points. All the points assigned by the
respondent must total 100; hence, the name of the scale ă constant sum.
 TOPIC 5 MEASUREMENT AND SCALING  91

Example:
In choosing a particular bath soap, indicate the importance you attach to each of
the following five aspects by distributing a total of 100 points among them.

Fragrance ________________
Colour ________________
Shape ________________
Texture of soap lather ________________
Total Points _______100______

5.3 NON-COMPARATIVE SCALING

TECHNIQUES
Non-comparative scales are often called monodic scales because only one
object is evaluated at a time, there is no comparison to another object or to
some specified ideal, such as „the perfect brand.‰ Respondents using a non-
comparative scale apply their own rating standard.

5.3.1 Continuous Scaling

A continuous rating scale allows the respondent to place a mark at any point
along a line running between two extreme points rather than selecting from
among a set of predetermined response categories.

Example:
How would you rate Parkson Grand as a department store?

Version 1
Probably the worst _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ Probably the best

Version 2
Probably the worst _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ Probably the best

0 10 20 30 40 50 60 70 80 90 100
92  TOPIC 5 MEASUREMENT AND SCALING

5.3.2 Itemised Rating Scales

A measurement scale having numbers and/or brief descriptions associated with
each category. The categories are ordered in terms of scale position.

Example:

1 2 3 4 5
Very unlikely Unlikely Neither unlikely nor likely Likely Very likely

1. I will be changing my job within the next 12 months

2. I will take on new assignments in the near future
3. It is possible that I will be out of this organisation within the next 12 months

(a) Likert Scale

A measurement scale with five response categories ranking from „strongly
agree‰ to „strongly disagree,‰ which requires the respondents to indicate a
degree of agreement or disagreement with each of a series of statements
related to the stimulus object.

Example:
Using the preceding Likert Scale, state the extent to which you agree
with each of the following statement:

My work is very interesting 1 2 3 4 5

I am not engrossed in my work at all 1 2 3 4 5
Life without work would be dull 1 2 3 4 5

(b) Semantic Differential Scale

A 7-point rating scale with end points associated with bipolar labels that have
semantic meaning. When using the semantic differential, the respondent is
typically asked to rate a brand, store or some other object in terms of these
bipolar adjectives such as cold and warm.
 TOPIC 5 MEASUREMENT AND SCALING  93

Example:
Responsive ______________ Unresponsive
Beautiful ______________ Ugly
Courageous ______________ Timid

(c) Stapel Scale

A scale for measuring attitudes that consists of a single adjective in the
middle of an even-numbered range of values.

Example:
State below how you would rate your supervisorÊs ability with respect to each
of the characteristics mentioned below, by circling the appropriate number.

+3 +3 +3
+2 +2 +2
+1 +1 +1
Adopting Modern Product Interpersonal
Technology Innovation Skills
ă1 ă1 ă1
ă2 ă2 ă2
ă3 ă3 ă3

5.3.3 Non-comparative Itemised Rating Scale Decision

One of the advantages of using non-comparative scales is the flexibility of
adapting them to a specific research project. When considering the itemised
rating scales, there are at least six factors that can be adjusted. These factors are
discussed below.

Numbers of scale categories to use: the number of categories should be between

five and nine; however, there is no single, optimal number of categories that
would be applicable for all scaling situations.

(a) Balanced versus Unbalanced

A balanced scale uses an equal number of favourable and unfavourable
categories; otherwise, the scale is unbalanced. The scale should be balanced
to obtain objective data. However, if the distribution of responses is likely
to be skewed, an unbalanced scale with more categories in the direction of
the skew may be appropriate.
94  TOPIC 5 MEASUREMENT AND SCALING

Example of a Balanced Scale:

Who decides which television programmes your children watch?

Child decides all of the time 

Child decides most of the time 
Child and parent decide together 
Parent decides most of the time 
Parent decides all of the time 
Example of an Unbalanced Scale:

Satisfied 
Neither satisfied nor dissatisfied 
Quite dissatisfied 
Very dissatisfied 
(b) Odd or Even Numbers of Categories
When there are an odd number of categories, the middle scale position
is generally designed as neutral. If a neutral or indifferent scale response
is a possibility for at least some of the respondents, an odd number of
categories should be use.

(c) Forced versus Non-forced Choice

A forced rating scale does not have a „no opinion‰ or „no knowledge‰
option. Thus, the respondents who do not have an opinion may mark
the middle scale position. If a sufficient number of the respondents, in
fact, do not have an opinion on the topic, marking the middle position
in this manner will distort measures of central tendency and variance. In
situations where the respondents are expected to have no opinion, the
accuracy of the data may be improved by having a non-forced rating scale
that includes a „no opinion‰ or „no knowledge‰ category.
 TOPIC 5 MEASUREMENT AND SCALING  95

Example:

How does CIMB Bank compare with Citibank?

CIMB Bank is better than Citibank 

CIMB Bank is about the same as Citibank 
CIMB Bank is worse than Citibank 
CanÊt say. 
(d) Nature and Degree of Verbal Descriptions
The strength of the adjectives used to anchor the scale has a slight
tendency to influence the distribution of the responses. With strong
anchors, respondents are less likely to use the extreme scale categories.

(e) Physical Form of the Scale

There is no agreement on which form is the most appropriate, but scales
could be presented vertically or horizontally; categories could be expressed
in terms of boxes, discrete lines or units on a continuum; categories may or
may not have numbers assigned to them; and numerical values could be
positive, negative or both.

5.3.4 Multi-items Scales

A multi-items scale consists of multiple items, where an item is a single question
or statement to be evaluated. Developing a multi-item rating scale requires
considerable technical expertise. The researcher begins by generating a set of
scale items based on theory, analysis of secondary data and qualitative research.

5.3.5 Scale Evaluation

What do you think are the scales that can be used for evaluation?

(a) Measurement Accuracy

Whenever items or individuals are measured, there is likely to be some
sort of error, whether intentional or unintentional. Intentional mistakes
may be seen in football team programmes, for example, when a player
is listed as 190 cm when he is actually 180 cm tall ă perhaps done to
96  TOPIC 5 MEASUREMENT AND SCALING

intimidate opponents. Unintentional mistakes may occur when something

under investigation is measured and the true response is sought but not
revealed. This is extremely common in research. Since virtually all research
efforts are flawed, marketing researchers must routinely measure the
accuracy of their information. Researchers must determine measurement
error, which is the difference between the information sought and the
information actually obtained in the measurement process.

(b) Reliability
Reliability refers to the ability of a scale to produce consistent results if
repeated measurements are taken. If a professor gives a group of students
two different (independent) tests to measure their knowledge of marketing
research and the studentÊs scores from the two measures are highly similar,
then the measures can be said to be reliable since they replicated each
otherÊs scores. Reliability is the extent to which scales are free of random
error and, thus, produce consistent results. In general, the less random error
detected, the more reliable the data will be. Systematic sources of error
do not negatively impact reliability, since they consistently influence the
measurement rather than create inconsistencies to it.

(c) Validity
Validity is the degree to which a test measures what it is supposed to
measure. In other words, the researcher should ask, „Are we measuring
what we think we are measuring?‰ For example, what if the marketing
exam had dealt with various aspects of marketing? Is this good enough to
make the exam on consumer behaviour valid? The answer is not always
obvious. All too often researchers think they are measuring one thing, when
they are actually measuring something else.

(d) Generalisability
Generalisability refers to each of the scales administered and their
interpretation in different research settings and situations. Hence, the
generalisability of a multi-item scale is determined by whether it can be
applied in a wide variety of data collection modes, whether it can be
used to obtain data from a wide variety of individuals, and under what
conditions it can be interpreted. As in the case of reliability and validity,
generalisability is not an absolute but rather, is a matter of degree.
 TOPIC 5 MEASUREMENT AND SCALING  97

CASE STUDY
Extreme Exposure Rock Climbing Center Faces the Krag
For the past five years, Extreme Exposure Rock Climbing Center has enjoyed a
monopoly. Located in Sacramento, California, Extreme Exposure was the dream
of Kyle Anderson who has been into freestyle extreme sports of various types
including outdoor rock climbing, hang gliding, skydiving, mountain biking,
snowboarding and a number of other adrenalin-pumping sports for many years.
Now in his mid-thirties, Kyle came to realise in the year of his thirtieth birthday
that, after three leg fractures, two broken arms and numerous dislocations, he
could not participate on the extreme edge like he used to. As a result, he found
an abandoned warehouse, recruited two investors and a friendly banker, and
opened up Extreme Exposure.

KyleÊs rock climbing centre has over 6,500 square feet of simulated rock walls to
climb, with about 100 different routes up to a maximum of 50 vertical feet or
about 15 metres. Extreme ExposureÊs design permits the four major climbing
types: top-roping, where the climber climbs up with a rope anchored at the top;
lead-climbing, where the climber tows the rope that he or she fixes to clips in the
wall while ascending; bouldering, where the climber has no rope but stays near
the ground; and rappelling, where the person descends quickly by sliding down
a rope. Climbers can buy day passes or monthly or yearly memberships. Shoes
and harnesses can be rented cheaply, and helmets are available free of charge as
all climbers must wear protective helmets. In addition to individual and group
climbing classes, Extreme Exposure has several group programmes including
birthday parties, a kidsÊ summer camp and corporate team-building classes.

Another rock climbing centre called the Krag will be built in Sacramento
within the next six months. Kyle notes the following items about the Krag that
are different from Extreme Exposure: (1) The Krag will have climbs up to a
maximum of 60 vertical feet (18 metres), (2) it will have a climber certification
programme, (3) there will be day trips to outdoor rock climbing areas, (4) there
will be group overnight and extended-stay rock climbing trips to the Canadian
Rockies, and (5) the KragÊs annual membership fee will be about 20 percent lower
than that of Extreme Exposure.
98  TOPIC 5 MEASUREMENT AND SCALING

Kyle chats with Dianne, one of his Extreme Exposure members who is in
marketing, during a break in one of her climbing visits, and Dianne summarises
what she believes Kyle needs to find out about his current members. DianneÊs list
is as follows:

1. What is the demographic and rock climbing profile of Extreme ExposureÊs

members?

2. How satisfied are the members with Extreme ExposureÊs climbing facilities?

3. How interested are its members in:

(a) Day trips to outdoor rock climbing areas;

(b) Group overnight and/or extended-stay rock climbing trips to the

Canadian Rockies; and

(c) A rock climber certification programme?

4. What are membersÊ opinions of the annual membership fee charged by

Extreme Exposure?

5. Will members consider leaving Extreme Exposure to join a new rock climbing
centre with climbs that are 10 feet higher than the maximum climb at Extreme
Exposure?

6. Will members consider leaving Extreme Exposure to join a new rock climbing
centre with climbs that are 10 feet higher than the maximum climb at Extreme
Exposure and whose annual membership fee is 20 percent lower than that of
Extreme Exposure?

EXERCISE 5.1

1. In a short essay, list and discuss the four primary scales

of measurement. Include an example of how each scale of
measurement is used in marketing research.
2. In a short essay, list and discuss the three types of comparative
scaling techniques. Include a specific example of each scaling
technique to support your answer.
3. In a short essay, list and discuss the three commonly used itemised
rating scales in marketing. Include at least one advantage and one
disadvantage of each itemised rating scale to support your answer.
 TOPIC 5 MEASUREMENT AND SCALING  99

 Measurement is an integrative process of determining the amount of

information about persons, events, concepts/ideas and/or objects of interest.
This means that researchers engage in using measurements by assigning
numbers or other symbols to characteristics of objects according to set rules.

 Scaling involves the generation of a continuum upon which measured objects

are located.

 The four basic scales of measurement are nominal, ordinal, interval and ratio.

 Of these, the nominal scale is the most basic. The numbers are used only for
identifying or classifying objects under study.

 The data classified are mutually exclusive and collectively exhaustive. In the
ordinal scale, the numbers indicate the relative position of the objects or
ability to order data, but not the magnitude of difference between them.

 The interval scale permits a comparison of the differences between the

objects. However, because it has an arbitrary zero point, it is not meaningful
to calculate ratios of scale values on an interval scale.

 The ratio scale, which has all the characteristics of the previously discussed
scales as well as a meaningful absolute zero point or origin, thus permitting
comparison of the absolute magnitude of the numbers and reflecting the
actual amount of the variable, is the highest level of measurement. The
researcher can compute ratios of scale values using this scale.

 Scaling techniques can be classified as comparative or non-comparative.

Comparative scaling involves a direct comparison of alternatives. Comparative
scales include paired comparisons, rank order and constant sum. The data
obtained through these procedures has only ordinal properties. Comparative
scaling can yield information vital to improving quality and determining
which attributes have the greatest impact on quality.

 In developed countries, respondents are quite used to providing responses on

interval and ratio scales. However, in developing countries, responses can be
best measured by using simple ordinal scales. Software is available to enable
researchers to construct a variety of comparative scales.
100  TOPIC 5 MEASUREMENT AND SCALING

 Non-comparative scales, also referred to as monadic or metric scales, refer to

objects which are scaled independently of each other. The resulting data is
generally assumed to be interval scaled.

 Non-comparative scales comprise continuous rating or itemised rating scales

(such as Likert, Stapel and semantic differential). Non-comparative scaling
is the most widely used scaling technique in marketing research. Ethical
considerations require that the appropriate type of scales be used in order to
get the data needed to answer the research questions and test the hypotheses.

Comparative scaling Non-comparative scaling

Interval ratio Scale evaluation
Nominal ordinal
Topic  Questionnaire
Design
6
LEARNING OUTCOMES
By the end of this topic, you should be able to:
1. Explain the purpose of a questionnaire and its objectives of asking
questions that the respondents can and will answer;
2. Describe the process of designing a questionnaire, the steps involved
and guidelines that must be followed in each step;
3. Construct questionnaires for specific research objectives using the
key elements discussed in the topic; and
4. Evaluate observational form of data collection.

 INTRODUCTION
This topic is concerned with designing the form on which data will be recorded.
This includes questionnaires, which are used to ask respondents questions, as
well as observation forms, which are designed to record respondentsÊ actions.
The researcher should be aware of the pitfalls in designing data collection
instruments. The choice of questions, words and even their order can be very
influential in affecting how respondents react to a survey. The researcher should
have an appreciation of all the preparation and thought that goes into the
creation of the final version. For this reason, this topic will describe the functions
of a questionnaire and the questionnaire development process. The topic also
discusses the dos and donÊts in question development and indicates the options
for questionnaire organisation. Finally, the topic also discusses the development
of the observational forms for data collection.
102  TOPIC 6 QUESTIONNAIRE DESIGN

6.1 THE IMPORTANCE OF QUESTIONNAIRES

In your opinion, why do we use questionnaires to collect data?

Data collection flows from specifying the information needed, to identifying the
method of obtaining it and to determining the procedures to be used. Once this
process is completed, a data collection instrument can be created. This is usually
in the form of questionnaires. It is vital that the questionnaire be constructed to:

(a) Capture the information needed;

(b) Be standardised so that the data is comparable across test sites and results
can thus be generalised; and

(c) Maximise the efficiency of tabulating and analysing data.

The questionnaires must translate the information needed into a set of specific
questions that the respondent can and will answer. Next, the questionnaires must
uplift, motivate and encourage the respondent to become involved in the
interviewing process. The questionnaire should also motivate the respondents to
cooperate and to complete the interview. Finally, a questionnaire should be
designed to minimise response error.
 TOPIC 6 QUESTIONNAIRE DESIGN  103

6.2 QUESTIONNAIRE DESIGN PROCESS

Figure 6.1: Questionnaire design process

Figure 6.1 above illustrates the steps involved in the questionnaire design
process.

6.2.1 Specify the Information Needed

The first step in questionnaire design involves translating the information needed
into a set of specific questions. Dummy tables are particularly helpful in guiding
individual question design and it is also important to have a clear idea of the
target population, so that the standard of the questionnaires designed suits the
target population. Therefore, the characteristics of the target population have a
great influence on questionnaire design. For instance, questionnaires appropriate
for school children may not be appropriate for college students.
104  TOPIC 6 QUESTIONNAIRE DESIGN

6.2.2 Type of Interviewing Method

In personal interviews, the respondents can see the questionnaires and they
can interact face-to-face with the interviewer. Therefore, it is best to use types
of questions that are lengthy, complex and diverse in a personal interview. In
a telephone interview, the questions asked are short and simple since the
respondents are asked to react to whatever they hear verbally over the telephone.
In mail interviews, the questions have to be simple, and detailed instructions
have to be provided because there is no interaction with the interviewer.

6.2.3 Individual Question Content

For each item of information that is needed, the researcher should consider
whether that information is being adequately obtained by questions already
formulated. Also, the data generated by each question should address the
information needed. Once it is determined that the question is needed, it is
important to make sure that the way questions are worded will obtain the desired
information. Sometimes, several questions are needed, instead of just one.

6.2.4 Overcoming Inability to Answer

Several factors that limit the respondentÊs ability to provide the desired
information are listed below:

(a) Are the Respondents Informed?

In situations where the respondents might not be informed about the topics
of interest, filter questions measuring familiarity, product use and past
experience should be asked before presenting questions on the topics.

(b) Can the Respondents Remember?

The ability to remember an event is influenced by the event itself, the time
elapsed since the event, and the presence or absence of events that would
assist in remembering.

(c) Can the Respondents Articulate?

Respondents should be provided with aids to help them articulate their
responses. For example, providing a list of descriptions or possible answers
can assist the respondents.
 TOPIC 6 QUESTIONNAIRE DESIGN  105

6.2.5 Overcoming Unwillingness to Answer

The reasons for unwillingness to respond include that too much effort is
required, the situation or context is seen as inappropriate for disclosure, no
legitimate purpose or need is seen for the information requested, and the
information is perceived as sensitive.

6.2.6 Choosing the Question Structure

Questions can be either structured or unstructured. Unstructured questions
are open-ended questions in which the respondents have the freedom to reply
in their own words rather than being limited to the selection of a response
alternative. These are best used as opening questions.

Structured questions pre-specify the set of response alternatives and the response
format. They improve respondent cooperation and make administering and
coding the survey easier. Structured questions can be divided into:

(a) Multiple-choice Questions

In multiple-choices questions, the researcher provides a choice of answers
and respondents are asked to select one or more of the alternatives given.

(b) Dichotomous Questions

A dichotomous question has only two response alternatives such as „yes‰
or „no‰, „agree‰ or „disagree.‰ Often, the two alternatives of interest are
supplemented by neutral alternatives such as „no opinion,‰ „donÊt know,‰
„both‰ or „none.‰

6.2.7 Choosing the Question Wording

This refers to the translation of the desired question content and structure
into words that can be clearly and easily understood by the respondent. The
following guidelines are used for framing questions:

(a) Define the Issue

A question should clearly define the issue being addressed.

(b) Use Ordinary Words

Ordinary words should be used in a questionnaire and they should match
the vocabulary level of respondents.
106  TOPIC 6 QUESTIONNAIRE DESIGN

(c) Avoid Ambiguous Words

The words used in a questionnaire should have a single meaning that is
known to the respondents.

(d) Avoid Leading or Biased Questions

A leading question is worded in such a way as to give a clue on how to
answer. Therefore, it can cause biased responses.

(e) Avoid Implicit Questions

An alternative that is not explicitly expressed in the options is an implicit
alternative. Making an implied alternative explicit may increase the
percentage of people selecting those alternatives.

(f) Avoid Implicit Assumptions

Questions should not be worded so that the answer is dependent upon
implicit assumptions about what will happen as a consequence. Implicit
assumptions are assumptions that are not stated in the questions.

(g) Avoid Generalisations and Estimates

Questions should be specific, not general. Moreover, questions should be
worded so that the respondent does not have to make generalisations or
compute estimates.

(h) Use Positive and Negative Statements

Many questions, particularly those measuring attitudes and lifestyles,
are worded as statements to which respondents indicate their degree of
agreement or disagreement. Evidence indicates that the response obtained
is influenced by the directionality or the statements, whether they are stated
positively or negatively.
 TOPIC 6 QUESTIONNAIRE DESIGN  107

6.2.8 Order of the Questions

The order in which the questions are presented is a crucial element in
questionnaire design because improper ordering of the questions can seriously
bias the responses. The opening questions should be interesting, simple and
non-threatening so that they will arouse the interest of the respondents and
involve them in the interviewing process. Questions that are difficult, sensitive,
embarrassing, complex or dull should be placed later in sequence only after
rapport has been established.

6.2.9 Form and Layout

It is good practice to divide a questionnaire into several parts, each part dealing
with a specific set of questions; research has shown that questions at the top of
the page receive more attention than those placed at the bottom.

6.2.10 Reproduction of the Questionnaires

In the reproduction of questionnaires, the following guidelines are used:

(a) The questionnaire should have a professional appearance;

(b) If the questionnaire runs for several pages, it should be presented in a

booklet form rather than a number of sheets that are paper clipped or
stapled together;

(c) An entire question should be reproduced on a single page;

(d) Vertical response category formats should be used for individual questions;

(e) The questions should not be overcrowded;

108  TOPIC 6 QUESTIONNAIRE DESIGN

(f) Directions or instructions meant for the interviewer or the respondent for
individual questions should be placed as close to the question as possible;
and

(g) Colour does not influence response rate.

Formats of question can be seen in APPENDIX 6A later in this topic.

6.2.11 Pre-testing
Pre-testing refers to testing the questionnaires on a small sample of respondents
to identify and eliminate potential problems, thereby improving the questionnaire.
In general, a questionnaire should not be used in a field survey unless it has been
adequately pre-tested.

SELF-CHECK 6.1

List down the steps involved in the questionnaire design process.

6.3 OBSERVATIONAL FORMS

Observational forms are designed to record respondent reaction to a new
product, advertising, packaging or some other marketing stimuli. Since there is
no questioning of the respondents, the researcher need not be concerned with the
psychological impact of the questions and the way they are asked. Observational
forms are designed primarily for the field work and the tabulation phases,
providing a guide for recording information accurately and simplifying the
coding, entry and analysis of data.

Observational forms should specify the who, what, where, why and how of
behaviour to be observed.
 TOPIC 6 QUESTIONNAIRE DESIGN  109

APPENDIX 6A
Formats of Question

1. Open-ended Question
Examples:
Do you think laws requiring passengers in motor vehicles to wear seat
belts are needed?
Do you intend to purchase an automobile this year?
Why did you purchase a Sony colour television set?

2. Multi-chotomous Question
Example:
Why did/would you purchase a Sony colour TV?
Price is lower than the other alternatives 
Feel it represents the highest quality 
Availability of local service 
Availability of service contract 
Picture is better 
Warranty is better 
Other 
3. Dichotomous Questions
Example 1:
Do you think laws requiring passengers in motor vehicles to wear seat
belts are needed?
_____________ Yes
_____________ No
110  TOPIC 6 QUESTIONNAIRE DESIGN

Example 2:
Do you intend to purchase an automobile this year?
_____________ Yes
_____________ No

Example 3:
Do you think that gasoline will be more or less expensive next year than
it is now?
_____________ More Expensive
_____________ Less Expensive

Do you think that gasoline will be less expensive or more expensive next
year than it is now?
_____________ More Expensive
_____________ Less Expensive

4. Scale
Example 1:
How often do you tape programmes for later viewing with your VCR?
_____________ Never
_____________ Occasionally
_____________ Sometimes
_____________ Often

Example 2:
Never Occasionally Sometimes Often

5. Ambiguous Words and Question

Example:
How often do you tape programmes for later viewing with your VCR?
_____________ Never
_____________ Occasionally
_____________ Sometimes
_____________ Often
 TOPIC 6 QUESTIONNAIRE DESIGN  111

6. Leading Question
Example:
Do you feel that limiting taxes by law is an effective way to stop the
government from picking your pocket on payday?
_____________ Yes
_____________ No
_____________ Undecided

7. Implicit Alternative
Example:
Would you like to have a job, if this were possible?

Would you prefer to have a job, or do you prefer to do just your house
work?

8. Double-barrelled Question
Example:

What is your evaluation of the price and convenience offered by XYZÊs

catalogue showroom?

CASE STUDY
Park Place Psychiatric Hospital
Park Place Hospital opened last year in Tucson, Arizona. It specialises in
psychiatric care and mental health services. Both inpatient and outpatient
services are provided, although the hospital is quite small and can only
accommodate up to 20 inpatients at any one time. Since the hospital is new and
its location is in the desert on the outskirts of Tucson, it has invested in an
extensive advertising campaign using billboards, newspaper and radio spots. By
the end of its first year of operation, Park Place has experienced only 45 percent
occupancy, but it is optimistic about the future.

The management of Park Place has decided that in order to grow, it must
reach out to its patient population in Tucson by being more aggressive in
its programme offerings. Among the services being considered is a series of
seminars on selected mental health care problems and a set of companion
programmes that will cover the various topics more extensively. The marketing
manager contacts a local research company and works with some of its personnel
to formulate a list of research objectives. These objectives address his concerns
112  TOPIC 6 QUESTIONNAIRE DESIGN

about the effectiveness of the marketing programme, the hospitalÊs location,

the decision-making process for a family member who detects a health-related
problem in another family member and an interest in various programmes and
seminars. These objectives include the following:

(a) To determine the level of interest in each of the following two-hour evening
seminars that cost $25 each: quitting smoking, weight control, stress
management, substance abuse, AlzheimerÊs disease, understanding anxiety
and coping with teenagers;

(b) To assess the degree of interest in enrolling in any of the two-month long
programmes previously listed that cost $250 each;

(c) To evaluate whether a person or family would seek help if a mental health
problem requiring professional counselling was evident;

(d) To determine if prospective clients recall Park Place Hospital advertising,

and if so, in which advertising medium;

(e) To evaluate the importance of location in the selection of a mental health

care facility for a family member, either as an inpatient or an outpatient;
and

(f) To obtain target market information.

Random digit dialling will be used to select prospective respondent households.

(a) Design a questionnaire suited for a telephone survey of 500 Tucson

households to be conducted with the „adult head of the household who is
responsible for the familyÊs health care.‰

(b) Justify your choice of the type of question response format for each
question. If you have used the same format for a group of related questions,
you should indicate your rationale for the group rather than for each
question in the group.

(c) Identify the organisation aspects of your questionnaire. Identify the

question flow approach you have used and indicate your reasons for
choosing it. Also, identify all screening questions, warm-ups, transition
questions and skip questions that you have used.
 TOPIC 6 QUESTIONNAIRE DESIGN  113

EXERCISE 6.1

In a short essay, list and discuss at least four guidelines that should be
followed in order to avoid problems in question wording.

Ć A researcher must design a questionnaire or an observation form in order

to collect quantitative primary data. A good questionnaire must fulfil three
objectives:
ă Translate the information needed into a set of specific questions;
ă Motivate respondents to complete the interview; and
ă Minimise response error.

Ć Designing a questionnaire is as much an art as it is a science. The text

can only provide the guidelines for development, but no one optimal
questionnaire design fits every research need.

Ć The questionnaire or research instrument must be adapted to the specific

cultural environment and should not be biased in terms of any one culture.
There are several software packages available in the market to facilitate
questionnaire design.

Observational forms Questionnaire pre-testing

Topic  Sampling
7
LEARNING OUTCOMES
By the end of this topic, you should be able to:
1. Identify the conditions that favour the use of a sample versus a
census;
2. Discuss the sampling design process;
3. Define the target population and determine the sample frame and
sample size for a specific study;
4. Select the sampling technique(s) appropriate for the study;
5. Evaluate the techniques as non-probability and probability sampling
techniques; and
6. Apply the probability and non-probability sampling techniques in
your own study.

 INTRODUCTION
International markets are measured in hundreds of millions of people; national
markets comprise millions of individuals; and even local markets may constitute
hundreds of thousands of households. To obtain information from every single
person in a market is usually impossible and obviously impractical. For these
reasons, marketing researchers make use of a sample. This topic describes how
researchers go about taking samples.
 TOPIC 7 SAMPLING  115

7.1 SAMPLING DESIGN AND PROCEDURES

The objective of marketing research is to obtain information about the
characteristics of a population, which involves using either a sample or a census.
A representative sample though not the entire population, contains the same
characteristics as the population; thus, generalisability is high, and population
parameters can be inferred from the sample information. This information is
contained in statistics, and the inferences that are made from it use statistical
techniques such as estimation procedures and hypothesis testing.

(a) A population is the aggregate of all the elements that share a common
set of characteristics and that contain the universe for the purpose of
the marketing research problem. The population parameters are typically
numbers, such as the proportion of consumers who are loyal to a particular
brand of toothpaste. Information about population parameters may be
obtained by taking either a census or a sample.

(b) A census involves a complete list of the elements of the population. The
population parameters can be calculated directly in a straightforward
manner after the census is enumerated.

(c) A sample, on the other hand, is a subgroup of the population selected for
participation in the study. Sample characteristics, called statistics, are then
used to make inferences about the population parameters.

Sampling is the fifth step in the process of formulating a research design. The
researcher has already specified the information needed to address the marketing
research problem, and the primary sources from which this information
would be obtained have also been determined. Furthermore, the scaling and
measurement procedures have been specified and the questionnaires designed.
At this point, sampling becomes critical because the researcher must now
administer the questionnaire to a relevant group of respondents.
116  TOPIC 7 SAMPLING

7.2 SAMPLING DESIGN PROCESS

What are the processes involved in designing a sample?

The sampling design process is depicted in Figure 7.1, which can thus serve as a
framework for discussion. The sampling design originates from the population of
interest; thus, the population must be accurately defined in a precise statement of
who should or should not be included in the sample. It may be very general, such
as residents of Sarawak, or specific such as Malay males, between 20 and 30 years
old, who cannot read. To measure the population, a sampling frame is needed.
The sampling frame consists of a list or set of directions that identifies the
elements of the target population; the sampling units must then be set. This is the
unit that contains the elements of the population to be sampled; for example, it
could be a household, an office or even an individual.

Figure 7.1: Sampling design process

At this point, the researcher is ready to introduce quantitative techniques into the
sampling design process. The appropriate sampling techniques must be chosen,
which involves making a decision between using a Bayesian approach versus a
traditional approach, sampling with replacement versus without replacement,
and using probability versus non-probability sampling. After the method is
determined, a particular sample size can be calculated. Thus, number represents
the total number or elements that must be surveyed for the results to be valid.
Finally, the execution and interviewing process must be operationalised (i.e.,
detailed specifications of variable definitions and interview procedures). This
step ensures that all interviewers use the same guidelines during the interview
process so that the data is comparable.
 TOPIC 7 SAMPLING  117

7.2.1 Define the Target Population

Sampling design begins by specifying the target population, which is the
collection of elements or objects that possess the information sought by the
researcher and about which inferences are to be made. The target population
must be defined precisely. Imprecise definition of the target population will
result in research that is ineffective at best and misleading at worst.

The target population should be defined in terms of elements, sampling

units, extent and time. An element is the object about which, or from which,
the information is desired. A sampling unit is an element, or a unit containing
the element, that is available for selection at some stage of the sampling process.

7.2.2 Determine a Sampling Frame

A sampling frame is a representation of the elements of the target population. It
consists of a list or set of directions for identifying the target population. Examples
of a sampling frame include the telephone book, an association directory listing
the firms in an industry, a mailing list purchased from commercial organisations,
a city directory or a map. If a list cannot be compiled, then at least some directions
for identifying the target population should be specified.

7.2.3 Select a Sampling Technique

The researcher must decide whether to use a Bayesian or traditional approach, to
sample with or without replacement, and to use non-probability or probability
sampling.

In the Bayesian approach, the elements are selected sequentially. After each
element is added to the sample, the data is collected, sample statistics computed
and sampling cost determined. The Bayesian approach explicitly incorporates
prior information about population parameters as well as cost and probabilities
associated with making wrong decisions.

In the traditional approach, the entire sample is selected before data collection
begins. Two approaches are available under traditional approaches, which are
sampling with replacement and sampling without replacement. In sampling with
replacement, an element is selected from the sampling frame and appropriate data is
obtained. Then, the element is placed back in the sampling frame. As a result, it is
possible for an element to be included in the sample more than once. In sampling
without replacement, once the element is selected for inclusion in the sample, it is
removed from the sampling frame and, therefore, cannot be selected again.
118  TOPIC 7 SAMPLING

7.2.4 Determine the Sample Size

Sample size refers to the number of elements to be included in the study.
Determining the sample size is complex and involves qualitative and quantitative
considerations. Important qualitative factors that should be considered in
determining the sample size include:

(a) The importance of the decision The more important the decision, the more
information is necessary and the more precisely this information should be
obtained.

(b) The nature of the research For exploratory research design, such as those
using qualitative research, the sample size is typically small; for conclusive
research, such as descriptive surveys, larger samples are required.

(c) The larger the number of variables involved, the larger the sample size.

(d) The nature of the analysis If sophisticated analysis of the data using
multivariate techniques is required, the sample size should be large. The
same applies if the data is to be analysed in great detail.

(e) Sample size used in similar studies Sample size is influenced by the
average size of samples in similar studies.

(f) Incidence rates This is the rate of occurrence of persons eligible to

participate in the study expressed as a percentage. The incidence rate
determines how many contacts need to be screened for a given sample size
requirement.

(g) Completion rates This refers to the percentage of qualified respondents

who complete the interview. It enables the researchers to take into account
anticipated refusals by people who qualify.

(h) Resource constraints In any marketing research project, money and time
are limited. Other constraints include the availability of qualified personnel
for data collection.

(i) The objectives of the research If the research objective is to gain insight,
the sample size need not be representative, but if the research objective is to
generalise the findings, a representative sample is needed.
 TOPIC 7 SAMPLING  119

7.2.5 Execute the Sampling Process

Execution of the sampling process requires a detailed specification of how the
sampling design decisions with respect to the population, sampling frame,
sampling unit, sampling techniques and sample size are to be implemented.

7.3 CLASSIFICATION OF SAMPLING

TECHNIQUES
Sampling techniques may be broadly classified as non-probability and probability
(see Figure 7.2). Non-probability sampling relies on the personal judgment of the
researcher, rather than chance, to select sample element; the researcher can
arbitrarily or consciously decide what elements to include in the sample. Non-
probability samples may yield good estimates of the population characteristics.
However, they do not allow for objective evaluation of the precision of the
sample results.

Figure 7.2: A classification of sampling techniques

120  TOPIC 7 SAMPLING

In probability sampling, sampling units are selected by chance. It is possible to

pre-specify every potential sample of a given size that could be drawn from the
population as well as the probability of selecting each sample. Every potential
sample need not have the same probability of selection, but it is possible to
specify the probability of selecting any particular sample of a given size. This not
only requires a precise definition of the target population, but also a general
specification of the sampling frame. Since sample elements are selected by
chance, it is possible to determine the precision of the sample estimates of the
characteristics of interest.

7.3.1 Non-probability Sampling Techniques

Non-probability sampling relies on the personal judgment of the researcher,
rather than chance, to select sample elements. The researcher can arbitrarily or
consciously decide what elements to include in the sample. Non-probability
samples may yield good estimates of the population characteristics.

(a) Convenience Sampling

This is a non-probability sampling technique that attempts to obtain
a sample of convenient elements. The selection of sampling units is left
primarily to the interviewer. Convenience sampling is the least expensive
and least time consuming of all sampling techniques. The sampling units
are accessible, easy to measure and cooperative. In spite of these advantages,
this form of sampling has serious limitations. Many potential sources of
selection bias are present, including respondent self-selection.

(b) Judgmental Sampling

This is a form of convenience sampling in which the population elements
are purposely selected based on the judgment of the researcher. Common
examples of judgmental sampling include:

(i) test market selected to determine the potential of a new product; and

(ii) purchase engineers selected in industrial marketing research because

they are considered to be representative of the company, etc.

Judgmental sampling is low cost, convenient and quick, yet it does not
allow direct generalisations to specific populations.
 TOPIC 7 SAMPLING  121

(c) Quota Sampling

This non-probability sampling technique is a two-stage restricted judgmental
sampling. The first stage consists of developing control categories or quotas
of population elements and in the second stage, sample elements are
selected based on convenience or judgment.

(d) Snowball Sampling

This technique is a non-probability sampling technique in which an initial
group of respondents is selected randomly. Subsequent respondents
are selected based on the referrals or information provided by the initial
respondents. This process may be carried out in waves by obtaining
referrals from referrals. A major objective of snowball sampling is to estimate
characteristics that are rare in the population. Examples include users of
particular government or social services.

7.3.2 Probability Sampling Techniques

A probability sampling technique is a sampling procedure in which each element
of the population has a fixed probabilistic chance of being selected for the
sample.

(a) Simple Random Sampling

This is a probability sampling technique in which each element in the
population has a known and equal probability of selection; every element
is selected independently of every other element and the sample is drawn
by a random procedure from a sample frame. To draw a simple random
sample, the researcher first compiles a sampling frame in which each
element is assigned a unique identification number. Then random numbers
are generated to determine which elements to include in the sample.

(b) Systematic Sampling

This sampling technique is a probability sampling technique in which the
sample is chosen by selecting a random starting point and then picking
every (n )th element in succession from the sampling frame.

(c) Stratified Sampling

This is a probability sampling technique that uses a two-step process to
divide the population into subpopulations or strata. Elements are selected
from each stratum by a random procedure. Technically, only simple
random sampling should be employed in selecting the elements from each
stratum.
122  TOPIC 7 SAMPLING

(d) Cluster Sampling

In this technique, the target population is first divided into mutually
exclusive and collectively exhaustive subpopulations, or clusters. Then, a
random sample of clusters is selected, based on a probability sampling
technique such as simple random sampling. For each selected cluster, either
all the elements are included in the sample or a sample of elements is
drawn probabilistically.

7.3.3 Choosing Non-probability versus Probability

Sampling
The choice between non-probability and probability samples should be based
on considerations such as the nature of the research, relative magnitude of
non-sampling versus sampling errors, variability in the population, as well as
statistical and operational considerations. For example, in exploratory research,
the findings are treated as preliminary and the use of the probability sampling
may not be warranted. On the other hand, in conclusive research, where the
researcher wishes to use the results to estimate overall market share or size of
the total market, probability sampling is favoured. Probability samples allow
statistical projection of the results to a target population.

For some research, highly accurate estimates of population characteristics are

required. In these situations, the elimination of selection bias and the ability
to calculate sampling error make probability sampling desirable. However,
probability sampling will not always produce more accurate results.

Another consideration is the homogeneity of the population with respect to the

variables of interest. A more heterogeneous population would favour probability
sampling because it would be more important to secure a representative sample.

7.4 INITIAL AND FINAL SAMPLE SIZE

DETERMINATION
This section focuses on the determination of sample size in simple random
sampling. We will define the various concepts and symbols and discuss the
properties of the sampling distribution.
 TOPIC 7 SAMPLING  123

7.4.1 Definition of Symbols

Confidence interval and other statistical concepts that play a central role in
sample size determination are defined in the following list:

(a) Parameter
A parameter is a summary description of a fixed characteristic or measure
of the target population. A parameter denotes the true value that would be
obtained if a census rather than a sample is undertaken.

(b) Statistic
Statistics is a summary description of characteristics or measure of the
sample. Sample statistics are used as an estimate of the population
parameter.

(c) Finite Population Correction

Finite population correction is a correction for overestimation of the
variance of a population parameter, for example, a mean of proportion,
when the sample size is 10 percent or more of the population size.

(d) Precision Level

When estimating a population parameter by using sample statistics,
the precision level is the desired size of the estimating interval. This is
the maximum permissible difference between the sample statistic and the
population parameter.

(e) Confidence Interval

The confidence interval is the range into which the true population
parameter will fall, assuming a given level of confidence.

(f) Confidence Level

The confidence level is the probability that a confidence interval will
include the population parameter.
124  TOPIC 7 SAMPLING

7.4.2 Sampling Distribution

ACTIVITY 7.1

In your opinion, what is the importance of sampling distribution?

The sampling distribution is the distribution of the values of a sample statistic

computed for each possible sample that could be drawn from the target
population under a specified sampling plan. Since many different samples could
be drawn from a population element, it is likely that the sample statistic derived
from any one sample will not equal the population parameter. As a result, the
sampling distribution supplies an approximation of the true value of population
parameters.

The main features of the sampling distribution include:

(a) It is normally distributed for large samples;

(b) The mean of the sampling distribution equals to m, the population

parameter for the mean;

(c) The standard error of the mean is the standard deviation of the sampling
distribution;

(d) Assuming no measurement error, the reliability of an estimate of a

population parameter can be assessed in terms of its standard error;

(e) The standard error of the mean can be estimated by using the sample
standard deviation, s, an estimator of s;

(f) Z-values calculate the area under the sampling distribution; and

(g) When the sample size is over 10 percent of the population size, the
standard error formulas overestimate the standard deviation of the
population parameter. A finite population correction factor is used to adjust
the estimates.
 TOPIC 7 SAMPLING  125

7.4.3 Statistical Approach to Determine Sample Size

In addition to statistical considerations, several qualitative factors should be
considered when determining the sample size. This includes the importance of
the decision, the nature of the research, the number of variables, the nature of the
analysis, sample sizes used in similar studies, incidence rates, completion rate
and resource constraints.

The statistical approach to determining sample size is based on traditional

statistical inference using formulas (equations). In this approach, the precision
level is specified in advance. This approach is based on the construction of
confidence intervals around sample means or proportion.

7.4.4 Confidence Interval Approach

The confidence interval approach to sample size determination is based on the
construction of confidence intervals around the sample means or proportions
using the standard error formula (see Figure 7.3).

Figure 7.3: Confidence interval approaches

For example, let us say that a researcher has taken a simple random sampling of
300 households to estimate the monthly expenses on department store shopping,
and has found that the mean monthly household expense for the sample is $182.
Past studies indicate that the population standard deviations can be assumed
to be $55. We want to find the interval within which a fixed proportion of the
sample means would fall. Suppose we want to determine an interval around the
population means that will include 95% of the sample means, based on samples
of 300 households. That is, we choose 95% as the confidence level. The 95% could
be divided into two equal parts, half below and half above the mean. Calculation
126  TOPIC 7 SAMPLING

of the confidence interval involves determining a distance below (XL) and above
(Xu) the population means (X), which contains a specified area of the normal
curve. The z-values corresponding to the XL and Xu values may be calculated as:

XL  
ZL 
X
Xu  
Zu 
X

Where ZL = z and Zu = +z, the lower value of X is:

Xu =  z  x

And the upper value of X is:

Xu =  z  x

Note that m is estimated by X. The confidence interval is given by:

X  z x
We can now set a 95 percent confidence interval around the sample mean of
4182. As the first step, we compute the standard error of the mean:

x   55 300  3.18
n

From the statistical tables, we can see that the central 95 percent of the normal
distribution lies within μ1.96 z values. The 95 percent confidence interval is given
by:
X μ 1.96 x
= 182.00 μ 1.96 (3.18)
= 182.00 μ 6.23

Thus, the 95 percent confidence interval ranges from 4175.77 to 4188.23. Thus,
the probability of finding the true population means is within the 4175.77 and
4188.23 percent. If the confidence interval is not reported, or cannot otherwise be
calculated from the information given, we have no way of knowing how precise
the sample estimates are.
 TOPIC 7 SAMPLING  127

(a) Sample Size Determination: Means

The confidence interval estimation formula can be used to determine the
sample size that will result in a desired confidence interval. In this case, we
use the same formula but have a different unknown, namely the sample
size, n. Suppose the researcher wants to estimate the monthly household
expense on department store shopping so that the estimate will be within
μ$500 of the true population value. What should the size of the sample be?
The following steps, will lead to an answer:

(i) Specify the level of precision. This is the maximum permissible

difference (D) between the sample mean and the population mean, in
our example, D = μ$500.

(ii) Specify the level of confidence. Suppose a 95 percent confidence level

is desired.

(iii) Determine the z-value associated with the confidence level using the
statistical table for 95 percent confidence level. In this case, the ability
that the population mean will fall outside one end of the interval is
0.025 (0.05/2) with an associated z-value of 1.96.

(iv) Determine the standard deviation of the population. The standard

deviation of the population may be available from a secondary
source, derived from estimates from pilot study data, or based on
the researcherÊs judgment. For example, the range of a normally-
distributed variable is approximately equal to, or plus or minus, three
standard deviations. The standard deviation of sample data can,
therefore, be estimated by dividing the range of that data by six. The
researcher can often estimate the range based on knowledge of
the phenomenon or knowledge of the scale used. For example, the
traditional Likert scale has a range of 4 (5 1).

(v) Determine the sample size using the formula for the standard error of
the mean.

(vi) If the population standard deviation, s, is unknown and an estimate is

used, it should be re-estimated once the sample has been drawn.

 2z 2
N 
D2
128  TOPIC 7 SAMPLING

(b) Sample Size Determination: Proportions

Proportions are estimated when the choice is dichotomous, meaning
there are only two categories (for example, yes or no). The researcher is
examining the percentage of elements that are in one of the two categories.
The researcher should follow these six steps:

(i) Specify the level of precision. Suppose the desired precision is such
that the allowable interval is set as D = P =  = μ0.05 (that is, 5 percent
expressed as the decimal equivalent).

(ii) Specify the level of confidence. Suppose that a 95 percent level is

desired.

(iii) Determine the z-value associated with the confidence level.

(iv) Estimate the population proportions, or . As explained earlier, the

population proportion may be estimated from secondary sources,
from the pilot study, or based on the researcherÊs judgment. Suppose
that based on secondary data, the researcher estimates that 64 percent
of the households in the target population possess a department store
credit card. In this case = 0.64.

(v) Determine the sample size using the formula of standard error of the
proportions. The derivation of the formula for sample size is similar to
that used in calculating means.

 (1   )z 2
N 
D2

7.4.5 Adjusting the Statistically Determined Sample

Size
(a) Incidence Rate
The rate represents the occurrence of persons eligible to participate in the
study expressed as a percentage.

(b) Completion Rate

The completion rate is the percentage of qualified respondents who
complete the interview. It enables researchers to take into account
anticipated refusals by people who qualify.
 TOPIC 7 SAMPLING  129

(c) Combined Adjustment

Combining the expected incident and completion rates results in a contacted
sample that is:
Final sample size
Initial sample size =
Incidence rate completion rate

7.4.6 Special Situations in Sample Size Determination

As a general rule, a small population situation is one in which the sample
exceeds 5 percent of the total population size. Notice that a small sample is
defined by the size of population under consideration. If the sample is less than
5 percent of the total population, you can consider the population to be of a large
size, and you can use the procedures described earlier in the topic.

On the other hand, if it is a small population, the sample size formula needs some
adjustment with what is called a finite multiplier, which is an adjustment factor
that is appropriately equal to the square root of that proportion of the population
not included in the sample.

For example, suppose our population size was considered to be 1,000 companies,
and we decided to take a sample of 500. That would result in a finite multiplier of
about 0.71, or the square root of 0.5, which is (1,000 0 500)/1,000). That is, we
could use a sample of only 355 (or .71 times 500) companies, and it would be just
as accurate as a sample of 500 if we have a large population. The formula of the
sample size using a finite multiplier is as follows:

N n
Small population sample size = Sample size formula X .
N 1

CASE STUDY
The Cobalt Group: Online Survey for MyCarToolsTm
The Cobalt Group (www.cobaltgroup.com) touts itself as a leading provider of
e-business products and services for the automotive industry. It claims almost
9,000 website clients and about the same number of part locator clients. There are
several Cobalt Group products, and a number are geared toward the automobile
dealer-automobile buyer/owner relationship. One of Cobalt GroupÊs products is
MycarToolsTM, which provides auto dealers with the ability to give personal
130  TOPIC 7 SAMPLING

web space to its auto buyers. This web space is set up to let the car owner input
his or her auto usage, to track service records, or to schedule a service appointment
online. The auto dealership can use this web space service as a direct-marketing
tool of its own, for as its customers enter their information on their web
spaces, the dealershipÊs database of owners grows. This database can be used to
identify customer types and trends, to target specific customer segments with
advertisements and special promotions, or even to alert customers to product
recalls.

There is great potential value in MycarToolsTM, for it is a means of maintaining

close contact with buyers of autos from the dealership over the life of their
automobiles. If it is satisfying to a customer, this contact may well translate into a
strong propensity for that customer to return to the dealership when he or she is
thinking about replacing the automobile. Of course, the degree of participation in
MycarToolsTM is entirely voluntary, and it depends on a number of factors.
First, the auto buyer must agree to be entered into the dealerÊs web space, and,
second, the buyer must have access to a computer and an Internet connection.
Finally, the buyer must use the service on a regular basis.

In an online problem-solving chat forum, automobile dealers who use the

Cobalt Group dealership software brainstormed questions about the value of
MycarToolsTM. The top question was, „Does the use of MycarToolsTM increase
the propensity of a car buyer to return to the dealership where he or she bought
the auto as compared to those who do not use MycarToolsTM?‰

A Cobalt Group Special Projects Team was assigned the task of researching
this question. The team immediately created an online questionnaire using
Websurveyor, but it had difficulty deciding on the sample plan.

1. What should the population definition for this marketing research situation
be?

2. Given this population definition, what should the sample frame be?

3. Should the Special Projects Team use an e-mail invitation or a website pop-up
invitation sampling method? Why did you recommend this method over the
other option?
 TOPIC 7 SAMPLING  131

EXERCISE 7.1

1. In a short essay, list and discuss the five steps in the sampling
design process.

2. Briefly discuss the difference between non-probability sampling

and probability sampling. Include a specific example of each type
of sampling to support your answer.

3. In a short essay, list and discuss at least four statistical concepts

used in sample size determination.

 Researchers can obtain information about the characteristics of a population

by conducting either a sample or a census. Sampling design begins by
defining the target population in terms of elements, sampling units, extent
and time. Then, the sampling frame should be determined. The sample size is
determined based on both quantitative and qualitative considerations.

 Sampling techniques may be classified as non-probability or probability

techniques. Non-probability sampling techniques rely on the researcherÊs
judgment; whereas in probability techniques, sampling units are selected by
chance.

 Companies focusing on customer value adopt appropriate sampling procedures

and techniques when collecting data for quality initiatives.

 The statistical approaches to determining sample size are based on confidence

intervals. These approaches may involve the estimation of the mean or
proportion.

 When estimating the mean or the proportion, determination of sample size

using the confidence interval approach requires the specification of precision
level, confidence level and population standard deviation.
132  TOPIC 7 SAMPLING

 When the sample size determined statistically represents the final or not
sample size that must be achieved, the initial sample size has to be much
larger to account for incidence rates and completion rates in order to deliver
the final sample size.

Non-probability Sampling design

Probability Symbol
Topic  Field Work
8
LEARNING OUTCOMES
By the end of this topic, you should be able to:
1. Explain the fieldwork process;
2. Discuss the training of field workers in making the initial contact,
asking questions, probing for answers, recording the answers and
terminating the interviews;
3. Analyse the supervision of field workers in terms of quality control
and editing, sampling control, control of cheating, and central office
control; and
4. Evaluate the field workers in areas of cost and time, response rates,
quality of interviewing and the quality of data.

 INTRODUCTION
Fieldwork is the fourth step in the marketing research process. It follows problem
definition, development of the approach, and formulation of the research design.
During this phase, field workers make contact with respondents, administer the
questionnaires or observation forms, record the data, and turn in the completed
forms for processing. A personal interviewer administering the questionnaires
door-to-door, an interviewer intercepting shoppers in a mall, a telephone
interviewer calling from a central location, a worker mailing questionnaires from
an office, an observer counting customers in a particular section of a store and
others involved in data collection and supervision of the process are all field
workers.
134  TOPIC 8 FIELD WORK

8.1 NATURE OF FIELD WORK

Marketing research data is rarely collected by the people who design the
research. Researchers have two major options for collecting their data. They can
develop their own resources within the organisation or they can employ a field
work agency. In either case, data collection involves the use of some kind of field
force. The field force may operate either in the field (personal in-home, mall
intercept, computer-assisted interviewing and observation) or from an office
(telephone, mail, e-mail and Internet survey). The field force who collect the data
typically have little research background or training. Ethical concerns are
particularly relevant to field work. Although there is ample opportunity for
violation of ethical standards, clients need not be overly concerned when dealing
with reputable field work agencies.

8.2 FIELD WORK/DATA COLLECTION PROCESS

What are the steps involved in collecting data?

All field work involves the selection, training and supervision of persons who
collect data (see Figure 8.1). The validation of field work and the evaluation of
field workers are also part of the data collection process.

Figure 8.1: Field work/data-collection process

Source: Malhotra, N. (2002). Basic marketing research: Application to
contemporary issues. Upper Saddle River, New Jersey, USA: Prentice-Hall, Inc.
 TOPIC 8 FIELD WORK  135

ACTIVITY 8.1

Provide examples of jobs that involve data collection.

8.3 SELECTION OF FIELD WORKERS

Several problems exist when selecting field workers:

(a) Finding workers who have the background characteristics that meet the job
requirement of a particular assignment can be a challenge. In particular, if
respondent and interviewer share several common characteristics, there is a
greater likelihood that the interview will be successful.

(b) Finding workers in various settings can be difficult if the research is being
conducted in different cities.

(c) It is also necessary to find researchers who possess the following key
attributes:
(i) Good health
(ii) An outgoing personality
(iii) Good communication skills
(iv) A pleasant appearance
(v) Adequate level of education
(vi) Experience in field work
136  TOPIC 8 FIELD WORK

8.4 TRAINING OF FIELD WORKERS

Training is essential to ensure the uniformity of data by minimising interviewer
and coding error. Important topics in interviewer training are illustrated in
Figure 8.2.

Figure 8.2: Training field workers

Source: Malhotra, N. (2002). Basic marketing research: Application to
contemporary issues. Upper Saddle River, New Jersey, USA: Prentice-Hall, Inc.

8.4.1 Making the Initial Contact

This step can bring about the successful cooperation of the respondent, who must
be convinced that his/her participation is important and that the survey has a
legitimate goal.

8.4.2 Asking Questions

Any slight change in the exact wording of a question can distort the meaning
intended by the researcher and the uniformity of the responses across the
sample.

The following are guidelines for asking questions:

(a) Be thoroughly familiar with the questionnaire;
(b) Ask the questions in the order in which they appear in the questionnaire;
(c) Use the exact wording given in the questionnaire;
(d) Read each question slowly;
(e) Repeat questions that are not understood;
(f) Ask every applicable question; and
(g) Follow instructions and skip pattern, probing carefully.
 TOPIC 8 FIELD WORK  137

8.4.3 Probing
Respondents are motivated to expand on or clarify a previous answer so the
interviewer can extract all information that is pertinent to the interview.

Commonly used probing techniques are as follows:

(a) Repeating the Question

Repeating the same question can be effective in eliciting a response.

(b) Repeating the RespondentÊs Reply

Respondents can be stimulated to provide further comments if the interviewer
repeats their replies verbatim. This can be done as the interviewer records the
replies.

(c) Using a Pause or Silent Probe

A silent probe, or an expectant pause or look, can cue the respondent
to provide a more complete response. However, the silence should not
become embarrassing.

(d) Boosting or Reassuring the Respondent

If the respondents hesitate, the interviewer should reassure the respondent
with comments, like, „there is no right or wrong answer. We are just trying
to get your opinion.‰ If the respondent needs an explanation of a word or
phrase, the interviewer should not offer an interpretation. Rather, the
responsibility for the interpretation should be returned to the respondent.
This can be done with a comment such as, „Just whatever it means to you.‰

8.4.4 Recording Answers

Precision in recording answers to both structured and unstructured responses is
critical to interpreting consumer responses accurately. Guidelines for recording
the answers are:
(a) Record responses during the interview;
(b) Use the respondentÊs own words;
(c) Do not summarise or paraphrase the respondentÊs answer;
(d) Include everything that pertains to the questionÊs objectives;
(e) Include all probes and comments; and
(f) Repeat the responses as they are written down.
138  TOPIC 8 FIELD WORK

8.4.5 Terminating the Interview

The respondents should be left with positive feelings about the interview,
and the research company should have all the information needed when the
interview is terminated.

In training, effort must be made to ensure interviewers:

(a) Understand the nature and content of questions;

(b) Understand how to record responses, code questionnaires and edit

interviews;

(c) Ensure respondentsÊ confidentiality;

(d) Ensure that the correct respondents are interviewed;

(e) Record a true picture;

(f) Execute their work clearly and accurately;

(g) Are prepared to deal with problem situations that may arise during
interviews;

(h) Are persuasive and able to minimise the number of selected households
and respondents who refuse to participate;

(i) Make quality a priority in all aspects of interviewing;

(j) Are courteous and friendly; and

(k) Strive for maximum efficiency without sacrificing quality.

8.5 SUPERVISION OF FIELD WORKERS

Supervision is needed to ensure that workers follow the procedures in which
they were trained, thus resulting in accurately completed surveys that will lead
to a meaningful analysis. It involves the following aspects (see Figure 8.3):
 TOPIC 8 FIELD WORK  139

Figure 8.3: Supervision of field workers

Source: Malhotra, N. (2002). Basic marketing research: Application
to contemporary issues. Upper Saddle River, New Jersey, USA: Prentice-Hall, Inc.

8.5.1 Quality Control and Editing

Quality control of field workers requires checking to see if the field procedures
are being properly implemented. If a problem is detected, the supervisor should
discuss it with the field workers and provide additional training if necessary.

8.5.2 Sampling Control

An important aspect of supervision is sampling control, which attempts to
ensure that the interviewers are strictly following the sampling plan rather
than selecting sampling units based on convenience or accessibility.

8.5.3 Control of Interviewer Cheating

Cheating involves falsifying a question or the entire questionnaire. An interviewer
may falsify part of an answer to make it acceptable or may provide fake answers.
140  TOPIC 8 FIELD WORK

8.5.4 Central Office Control

The supervisor provides quality and cost-control information to the central office
so that a total progress report can be maintained. In addition to the control
initiated in the field, other controls may be added at the central office to identify
potential problems.

8.6 VALIDATION OF FIELD WORKERS

Validation of fieldwork means verifying that the field workers are submitting
authentic interviews. To validate the study, the supervisors call 10 to 25 percent
of the respondents to inquire whether the field workers actually conducted the
interview. The supervisor asks about the length and quality of the interview, or
the respondents' reaction to the interviewer, and collects basic demographic data.

8.7 EVALUATION OF FIELD WORKERS

Field workers must be evaluated so those who meet objectives are identified. As
a result, a better field force can be built, and poorer workers can be trained
to improve performance. Workers should also be evaluated in terms of time,
response rates, quality of interviewing and quality of data (see Figure 8.4).

Figure 8.4: Evaluating field workers

Source: Malhotra, N. (2002). Basic marketing research: Application
to contemporary issues. Upper Saddle River, New Jersey, USA: Prentice-Hall, Inc.
 TOPIC 8 FIELD WORK  141

The list of factors upon which the field workers should be evaluated is as follows:

(a) Cost and Time

(i) Total cost per completed interview.
(ii) Total time per completed interview.

(b) Response Rate

(i) Percentage of completed interview over contacts.

(c) Quality of Interviewing

(i) Appropriateness of the introduction.
(ii) Precision of asking questions.
(iii) Ability to probe without bias.
(iv) Ability to ask sensitive questions.
(v) Interpersonal skills.
(vi) Manner of terminating the interview.

(d) Quality of Data

(i) Recorded legibly.
(ii) Followed instructions.
(iii) Verbatim recording of unstructured question responses.
(iv) Meaningful recording of unstructured question responses.
(v) Low incidence of item non-response.

EXERCISE 8.1

1. Briefly discuss the process of selecting field workers.

2. In a short essay, discuss the importance of properly evaluating

field workers and discuss five elements that are important when
conducting these evaluations.
142  TOPIC 8 FIELD WORK

 Researchers have two major options for collecting data: developing their own
field workers within the organisation or employing field work agencies. In
either case, data collection involves the use of a field force.

 A field force should be healthy, outgoing, communicative, pleasant, educated

and experienced. They should be trained in important aspects of fieldwork,
including making initial contact, asking questions, probing for answers,
recording answers and terminating the interview.

 Supervision of field workers involves quality control and editing, sampling

control, control of cheating and central office control. Validation of fieldwork
can be accomplished by calling 10 to 25 percent of those who have been
identified as interviewees and inquiring whether the interviews took place.

 Field workers should be evaluated on the basis of cost and time, response
rates, quality of interviewing and quality of data collection.

Field work Recording answers

Interviewer training Sampling control
Probing
Topic  Data
Preparation
9 and Analysis
Strategy
LEARNING OUTCOMES
By the end of this topic, you should be able to:
1. Discuss the nature and scope of the data preparation process;
2. Describe questionnaire checking and editing activities and treatment
of unsatisfactory responses;
3. Apply the guidelines for coding questionnaires;
4. Review the data cleaning process and methods used to treat missing
responses; and
5. Appraise the procedures for selecting a data analysis strategy and
factors influencing the process.

 INTRODUCTION
Data preparation and analysis is the fifth step of the marketing research process.
In this topic, we discuss data preparation and analysis strategy that emphasises
their importance. Issues involved in questionnaire checking, coding, transcribing,
and data cleaning, emphasising the treatment of missing responses, are discussed.
The selection of data analysis strategy and classification of statistical techniques
are also presented.
144  TOPIC 9 DATA PREPARATION AND ANALYSIS STRATEGY

9.1 DATA PREPARATION PROCESS

Do you know what a data preparation process is? Figure 9.1 shows the data
preparation process for primary data collection involving questionnaires. The
entire process is guided by the preliminary plan of data analysis that was
formulated in the research design phase. The first step in a data preparation
process is when an acceptable questionnaire is checked by a fieldwork supervisor.
This is followed by editing, coding and transcribing the data, all done by the data
collection agency. The researcher cleans the data then performs further checks for
consistency and specifies how missing responses will be treated. The raw data
must be subjected to a statistical analysis to render it usable, which will result
in conclusions about the study. The quality of the statistical inferences may be
compromised if adequate attention is not paid to the data preparation phase,
which may result in biased findings and incorrect interpretation. If any problems
are detected, the fieldwork can be modified to incorporate corrective action.

Preliminary Plan of Data Analysis

Questionnaire Checking

Editing

Coding

Transcribing

Data Cleaning

Selecting a Data Analysis Strategy

Figure 9.1: Data preparation process

Source: Malhotra, N. (2002). Basic marketing research: Application
to contemporary issues. Upper Saddle River, New Jersey, USA: Prentice-Hall, Inc.
 TOPIC 9 DATA PREPARATION AND ANALYSIS STRATEGY  145

SELF-CHECK 9.1

Identify the steps in the data preparation process.

9.2 QUESTIONNAIRE CHECKING

Why do we need to check the questionnaire? The questionnaire must be examined
to ensure that it will obtain the desired information. If a questionnaire fails to
adequately gather the necessary information, the results of the study will be
incomplete and possibly misleading. Questionnaire checking is done in two stages:
(a) Before the field work, checking is done to ensure completeness and quality.
(b) After the field work, checking is done to identify:
(i) Partially answered questionnaires;
(ii) Physically incomplete questionnaires (missing pages);
(iii) Pattern of responses showing little variance;
(iv) Questionnaires received after the cut-off date; and
(v) Questionnaires answered by inappropriate respondents.

If quotas or cell group sizes have been imposed, the acceptable questionnaire
should be classified and counted accordingly. Any problems in meeting the
sampling requirements should be identified and corrective action taken.

9.3 EDITING
Do you know what is involved in editing? Editing is a review of the
questionnaires with the objective of increasing accuracy and precision. It consists
of screening questionnaires to identify illegible, incomplete, inconsistent or
ambiguous responses. The treatment of unsatisfactory responses is shown in
Figure 9.2. Unsatisfactory responses discovered in editing are treated in any of
the following ways, depending on the particular situation:

(a) Returning to the Field

The interviewer re-contacts the respondents who answered the
questionnaires. However, the data obtained may not be identical to the
original data due to changes over time and difference in the mode of
146  TOPIC 9 DATA PREPARATION AND ANALYSIS STRATEGY

questionnaire administration. This approach is common in the case of

industrial and business marketing surveys where the sample sizes are small
and the respondents are identifiable.

(b) Assign Missing Value

With this method, the researcher simply assigns a missing value to the
unsatisfactory responses because of the difficulties involved in re-
contacting the respondents. This is preferred if both the number of erring
respondents and the proportion of such errors for each respondent are
small and the affected variables are not key ones.

(c) Discard Unsatisfactory Respondents

This approach is followed when the sample size is large, the proportion
of respondents who have erred is low (less than 10%), the proportion
of unsatisfactory responses for each respondent is large, the erring
respondents do not differ from the satisfactory respondents in terms of
demography or product usage characteristics and the key variables are
affected in the process. However, the decision to designate a respondent as
unsatisfactory is subjective and so the procedure adopted to identify such
respondents along with their number should be reported.

Unsatisfactory respondents may differ from satisfactory respondents in a

systematic way and the decision to designate a respondent as unsatisfactory
may be subjective.

These factors related to unsatisfactory respondents may bias the results. If

the researcher decides to discard unsatisfactory respondents, the procedure
adopted to identify these respondents and the number of discarded
respondents should be reported by the researcher in the project report.
 TOPIC 9 DATA PREPARATION AND ANALYSIS STRATEGY  147

Treatment of
Unsatisfactory
Responses

Return to the Field Assign Missing Discard

Values Unsatisfactory
Respondents

Substitute a Casewise Deletion Pairwise Deletion

Neutral Value

Figure 9.2: Treatment of unsatisfactory responses

Source: Malhotra, N. (2002). Basic marketing research: Application
to contemporary issues. Upper Saddle River, New Jersey, USA: Prentice-Hall, Inc.

SELF-CHECK 9.2

Explain how the unsatisfactory responses discovered in editing are

treated.

9.4 CODING
Do you know what coding is? Coding means the assignment of a code to a
specific response to a specific question along with the data record and column
position that code will occupy. Coding occurs after editing and before
transcribing. For unstructured questions, a variety of answers are possible.
Therefore, the developing of codes usually follows data collection.
148  TOPIC 9 DATA PREPARATION AND ANALYSIS STRATEGY

For example, sex of respondents may be coded as 1 for females and 2 for males. A
field represents a single item of a data such as sex of respondent, marital status,
age and so on. Each record can have several columns.

Generally, all the data will be stored on a single record although a number of
records may be used for each respondent. This is helpful to prepare codebooks
containing the coding instructions and the necessary information about the
variables in the data set. Questionnaire coding becomes simpler if the code book
is used because it helps the researcher identify and locate variables.

All the records pertaining to all the respondents are stored in a computer file.
One can also use a spreadsheet programme to enter the data, as most analysis
programmes can import data from spreadsheets.

If the questionnaire contains only structured questions or very few unstructured

questions, it is per-code. This means that codes are assigned before fieldwork
is conducted. If the questionnaire contains unstructured questions, codes are
assigned after the questionnaires have been returned from the field (post-coding).

9.4.1 Coding Questions

The respondent code and the record number should appear on each record in the
data. The following additional codes should be included for each respondent:
(a) Project code;
(b) Interview code;
(c) Date and time codes; and
(d) Validation code.

A fixed field code is a code in which the numbers of records for each respondent
are the same and the same data appears in the same columns for all respondents.
Fixed field codes are highly desirable. If possible, the standard codes should be
used for missing data. The missing value codes should be distinct from the codes
assigned to the legitimate responses.
 TOPIC 9 DATA PREPARATION AND ANALYSIS STRATEGY  149

Coding structured questions is relatively simple. The researcher assigns a code

for each response to each question and specifies the appropriate record and
columns in which the response codes are to appear. For example, to a question, a
„Yes‰ response is coded as 1 and a „No‰ response coded as 2. Generally, a single
column is sufficient to code a structured question with a single response if there
are fewer than nine possible responses.

For questions that permit multiple responses, each possible response option
should be assigned a separate column. For example, the questions include those
about ownership or usage, television viewing and magazine readership.

Coding for unstructured or open-ended questions is more complex. RespondentsÊ

verbatim responses are recorded on the questionnaire. Codes are developed and
assigned to these responses. The following guidelines are suggested for coding
unstructured or open-ended questions:

(a) The category codes determined should be mutually exclusive and collectively
exhaustive.

Categories are mutually exclusive if each response fits into one and only
one category code. Ideally, categories should not overlap.

Categories are collectively exhaustive if every response fits into one of the
assigned category codes. It can be achieved by adding a category code of
„other‰ or „none of the above‰.

(b) Data should be coded to retain as much detail as possible.

Obtaining information on the exact number allows the researcher to define
the categories. If the categories were predefined, the subsequent analysis of
data would be limited by those categories.
150  TOPIC 9 DATA PREPARATION AND ANALYSIS STRATEGY

9.5 TRANSCRIBING
What is transcribing? Transcribing data involves transferring the coded data
from the questionnaires or coding sheets into disks or directly into computers by
key punching. This step is unnecessary if the data is entered directly into the
computer, as it is collected via computer-assisted telephone interviewing (CATI)
or personal interviewing (CAPI). Besides key punching, the data can be transferred
by using mark sense forms, optical scanning or computerised sensory analysis
(see Figure 9.3).

Raw Data

CATI / Key Punching Mark Optical Computerised

CAPI via CRT Sense Scanning Sensory
Terminal Form Analysis

Verification: Correct
Key Punching Error

Computer Magnetic
Memory Disks
Tapes

Transcribed Data

Figure 9.3: Data transcription

Source: Malhotra, N. (2002). Basic marketing research: Application
to contemporary issues. Upper Saddle River, New Jersey,
USA: Prentice-Hall, Inc.
 TOPIC 9 DATA PREPARATION AND ANALYSIS STRATEGY  151

When CATI, CAPI, or other electronic methods are used, data is verified as it is
collected. In the case of inadmissible responses, the computer will prompt the
interviewer or respondent. In case of admissible responses, the interviewer or the
respondent can see the recorded response on the screen and verify it before
proceeding.

The selection of a data transcribing method depends on the type of interviewing

methods used and the availability of equipment.

9.6 DATA CLEANING

Data cleaning occurs after transcribing but before statistically adjusting the data.
Cleaning the data refers to a thorough check performed on the data to ensure its
reliability. The procedure requires consistency checks and treatment of missing
responses.

9.6.1 Consistency Checks

Consistency checking is part of the data cleaning process that identifies:

(a) Data that is Out-of-range

Out-of-range data values are inadmissible and must be corrected. These
out-of-range responses can arise due to respondent or interviewer errors.
Computer software packages such as SPSS, SAS, MINITAB and MSExcel
can be programmed to identify out-of-range values. This makes it easy
to systematically check each variable for out-of-range values. The correct
responses can be determined by going back to the edited and coded
questionnaire.

(b) Data that is Logically Inconsistent

Responses can be logically inconsistent in various ways. The necessary
information (respondent code, variable code, variable name, record
number, column number and inconsistent values) can be printed to locate
these responses and take corrective action.

(c) Data that has Extreme Values

Extreme values should be closely examined. Not all extreme values result
from errors but they may point to problems with the data.
152  TOPIC 9 DATA PREPARATION AND ANALYSIS STRATEGY

9.6.2 Treatment of Missing Responses

Missing responses refer to values of a variable that are unknown as these
respondents did not provide unambiguous answers to the question. Missing
responses would pose a problem if their proportion to the total is significant
(more than 10 percent). Available strategies for alleviating the missing data
problem include:

(a) Substitute a Neutral Value

A neutral value mean response to the variable is substituted for missing
responses. The mean of the variables remains unchanged and other
statistics, such as correlations, may not be affected much if the proportion
of the missing value is small.

(b) Casewise Deletion

In casewise deletion, cases or respondents with any missing responses
are discarded from the analysis. If several variables are involved in the
analysis, respondents with missing values on any of those variables will be
excluded. Throwing away large amounts of data is undesirable because it is
costly and time-consuming to collect data. Furthermore, respondents with
missing responses could differ from respondents with complete responses
in systematic ways. If so, casewise deletion could seriously bias the result.

(c) Pairwise Deletion

Pairwise deletion is a method of handling missing values in which for
each calculation or analysis, only the cases or respondents with complete
responses are considered. The researcher is interested in calculating
pairwise correlations between a set of variables.

In calculating the correlation between attitude and preference, all respondents

with legitimate responses with two variables will be included in the analysis.
This is done even if these respondents may have missing values on the
other two variables. Pairwise deletion may be appropriate when:
(i) The sample size is large;
(ii) There are few missing responses; and
(iii) Variables are not highly related.

This procedure can produce results that are unappealing or even infeasible,
as different calculations in an analysis may be based on different sample
sizes.
 TOPIC 9 DATA PREPARATION AND ANALYSIS STRATEGY  153

The different procedures for the treatment of missing responses may yield
different results, particularly when the responses are not missing at random and
the variables are related. Researchers should carefully consider the implications
of the various procedures before selecting a particular method for the treatment
of non-response.

9.7 SELECTING A DATA ANALYSIS STRATEGY

Figure 9.4 shows the steps in selecting a data analysis strategy.

Earlier Steps of the Marketing Research Process

Known Characteristics of Data

Properties of Statistical Technique

Background and Philosophy of the Researcher

Data Analysis Strategy

Figure 9.4: Selecting a data analysis strategy

Source: Malhotra, N. (2002). Basic marketing research: Application
to contemporary issues. Upper Saddle River, New Jersey, USA: Prentice-Hall, Inc., p.461.

Let us now look at the explanation of each step.

(a) Earlier Steps of the Marketing Research Process

When selecting a data analysis strategy, the researcher must begin by
considering the early steps in the process:
(i) Problem definition;
(ii) Development of an approach; and
(iii) Research design.
154  TOPIC 9 DATA PREPARATION AND ANALYSIS STRATEGY

(b) Known Characteristics of the Data

The second step is to consider the known characteristics of the data. The
scales of measurement (nominal, ordinal, interval or ratio) and the nature of
the design strongly influence the choice of a statistical technique.

For Example:

(i) Categorical scale versus Categorical scale: Use chi-square test

(e.g. race versus income).

(ii) Categorical (two scales) versus Continuous Scale: Use t-student test
(e.g. gender versus attitudinal statements).

(iii) Categorical (more than two scales) versus Continuous Scale: Use
F-test (e.g. marital status versus attitudinal statements).

(c) Properties of Statistical Techniques

In the third step, it is important to take into account the properties of
statistical techniques, particularly their purpose and underlying assumptions.
Thus, depending on the applications (e.g. examining difference in the
variables, making predictions) appropriate statistical techniques should be
chosen. The statistical techniques may be broadly classified as univariate or
multivariate.

(i) Univariate Techniques

Used for analysing data when there is a single measurement of each
element in the sample, or if there are several measurements of each
element, each variable is analysed in isolation. For example, number
of male respondents versus number of female respondents.

(ii) Bivariate and Multivariate Techniques

Used for analysing data when there are two or more measurements
on each element and the variables are analysed simultaneously. This
technique is concerned with the simultaneous relationships among
two or more phenomena.

(d) Background and Philosophy of the Researcher

The last step, which is the researcherÊs background and philosophy, affects
the choice of data analysis strategy. Depending on their sophistication,
researchers employ various statistical methods and make different
assumptions about the variables and underlying population. The experienced,
statistically-trained researcher will employ a range of techniques.
 TOPIC 9 DATA PREPARATION AND ANALYSIS STRATEGY  155

9.8 STRATEGIES FOR QUALITATIVE DATA

In providing an overview of qualitative analysis, it is apparent that there are
different approaches to the commencement of this process. These relate to
whether you start from a deductive or an inductive perspective. If you begin
your research project from a deductive position you will seek to use existing
theory to shape the approach which you adopt to the qualitative research process
and to the aspects of data analysis. On the other hand, if you commence your
research project from an inductive position, you will seek to build up theory
which is adequately grounded in a number of relevant cases. The design of
qualitative research requires you to recognise this choice and to devise an
appropriate strategy to guide your research project.

Yin (1994) has identified analytical strategies related to these two approaches
which may be used to analyse qualitative research:

(a) Using a Theoretical or Descriptive Framework to Analyse Data

Yin (1994) suggests that where an existing theory has been used to
formulate the research question and objectives, the same theoretical
propositions should also be used as a means to devise a framework to help
organise and direct data analysis. This approach demonstrates a preference
for commencing with and utilising theory in qualitative research, rather
than allowing it to develop from the work.

(b) Exploring Qualitative Data Without a Predetermined Theoretical or

Descriptive Framework
The alternative analytical strategy to the previous approach is to start to
collect the data then to explore it to see which themes or issues to follow up
and to concentrate on (Glaser and Strauss, 1967; Robson, 1993; Schatzman
and Strauss, 1973; Strauss and Corbin, 1998; Yin, 1994). Yin (1994) believes
that this may be a difficult strategy to follow and may not lead to success
for someone who is an inexperienced researcher. Robson (1993) agrees that
this is likely to be the case where a researcher simply goes ahead and
collects data without examining it to assess which themes are emerging
from the data being gathered. However, to use this strategy successfully
is likely to involve a lengthy period of time and it would be resource
intensive.
156  TOPIC 9 DATA PREPARATION AND ANALYSIS STRATEGY

EXERCISE 9.1

Tick „True‰ or „False‰ in the appropriate boxes.

No Statements True False

1. Unsatisfactory response from collected data may
be desirable when the proportion of unsatisfactory
respondents is between 20 and 30 percent.
2. Deleting the question from the survey is an alternative
for the treatment of unsatisfactory responses.
3. The assignment of a symbol to represent a specific
response to a specific question, along with the data
record and column position that symbol will occupy is
called coding.
4. Transcribing data involves transferring the coded data
from the questionnaires or coding sheets onto disks
directly into a computer by key punching or other
means.
5. The thorough and extensive check for consistency and
treatment of missing responses is called validating.
6. A part of the data cleaning process that identifies data
that is out of range or logically inconsistent, or that has
extreme values, is called a consistency check.
7. A method of handling missing responses in which
respondents with any missing responses are discarded
from the analysis is called pairwise deletion.
8. Statistical techniques appropriate for analysing data
when there is a single measurement of each element in
the sample are called univariate techniques.
9. Multivariate statistical techniques are appropriate for
analysing data if there are several measurements of each
element and each variable is analysed in isolation.
10. Univariate techniques are statistical techniques
suitable for analysing data when there are two or more
measurements on each element and the variables are
analysed simultaneously.
 TOPIC 9 DATA PREPARATION AND ANALYSIS STRATEGY  157

EXERCISE 9.2

Multiple Choice Questions

1. Which of the following is the last stage of the data preparation

process?
A. Questionnaire checking
B. Coding
C. Data cleaning
D. Selecting a data analysis strategy

2. _____________ consists of screening questionnaires to identify

illegible, incomplete, inconsistent or ambiguous responses.
A. Coding
B. Editing
C. Validating
D. Pre-testing

3. Which of the following is NOT an alternative for the treatment of

unsatisfactory responses?
A. Returning the questionnaire to the field to get better data
B. Assigning missing values
C. Discarding unsatisfactory respondents
D. Deleting the question from the survey

4. Discarding unsatisfactory responses from collected data may

have merit in all of the following situations EXCEPT:
A. When the proportion of unsatisfactory respondents is small
B. When the sample size is large
C. When unsatisfactory respondents differ from satisfactory
respondents in obvious ways
D. When the proportion of unsatisfactory responses for each
of these respondents is large
158  TOPIC 9 DATA PREPARATION AND ANALYSIS STRATEGY

5. Categories are_____________ if each response fits into one and

only one category code.
A. dependent
B. mutually inclusive
C. collectively exhaustive
D. mutually exclusive

6. A part of the data cleaning process that identifies data that are out
of range or logically inconsistent, or that have extreme values is
called _____________.
A. transcribing
B. validating
C. consistency checks
D. coding

7. Which of the following is NOT an option available for the

treatment of missing responses?
A. Substitute a neutral value
B. Casewise deletion
C. Modify the questionnaire
D. Pairwise deletion

8. Pairwise deletion may be appropriate when all of the following

conditions are present EXCEPT:
A. The sample size is large
B. There are few missing responses
C. The variables are not highly related
D. The sample size is small
 TOPIC 9 DATA PREPARATION AND ANALYSIS STRATEGY  159

9. Statistical techniques appropriate for analysing data when there is a

single measurement of each element in the sample are called
_____________.
A. Univariate techniques
B. Multivariate techniques
C. Random techniques
D. Parallel techniques

10. Which of the following statistical techniques is appropriate for

analysing data if there are several measurements of each element and
each variable is in isolation?
A. Univariate techniques
B. Multivariate techniques
C. Random techniques
D. Parallel techniques

 Data preparation begins with a preliminary check of all questionnaires

for completeness and interviewing quality. Then, editing takes place.
Editing consists of screening questionnaires to identify illegible, incomplete,
inconsistent or ambiguous responses. The next step is coding.

 A numerical or alphanumeric code is assigned to represent a specific

response to a specific question, along with the column position that code will
occupy. It is helpful to prepare a codebook.

 The coded data is transcribed onto disks or entered into computers via
keypunching. Besides keypunching, the data can be transferred by using
mark sense forms, optical scanning or computerised sensory analysis.

 Cleaning the data requires consistency checks and treatment of missing

responses. Options available for treating missing responses include
substitution of a neutral value such as the mean, casewise deletion and
pairwise deletion.
160  TOPIC 9 DATA PREPARATION AND ANALYSIS STRATEGY

 The selection of the data analysis strategy should be based on the earlier
steps of the marketing research process, known characteristics of the data,
properties of statistical techniques and the background and philosophy of the
researcher. Proper data preparation is essential before analysing data for
quality management programmes.

Cleaning data Questionnaire

Editing data Transcribing
Coding data
Topic  Report
Preparation
10 and
Presentation
LEARNING OUTCOMES
By the end of this topic, you should be able to:
1. Outline the basic requirements of report preparation including report
format, report writing, graphs and tables;
2. Examine the approaches used in the marketing research report from
the clientÊs perspective and explain the guidelines for reading the
research report;
3. Discuss the reasons for follow-up with the client;
4. Summarise the assistance that should be given to the client and the
evaluation of the research project; and
5. Write a research report and present it.

 INTRODUCTION
Report preparation and presentation constitutes the sixth and final step of
the marketing research process. It follows problem definition, developing
an approach, research design formulation, field work and data preparation
and analysis. Guidelines for report preparation, including report writing and
preparing tables and graphs are provided in this section. Oral presentation of the
report will also be discussed.
162  TOPIC 10 REPORT PREPARATION AND PRESENTATION

10.1 IMPORTANCE OF THE REPORT AND

PRESENTATION
Why is it important to have a report and presentation? The report and its
presentation are tangible products of the research effort, and the report serves as
a historical record of the project. If inadequate attention is paid to this step, the
value of the project to management will be greatly diminished.

The manager evaluates the quality of the entire project based on the quality of
the report and presentation. ManagementÊs decision to undertake marketing
research in the future or to use the particular research suppliers again will be
influenced by the perceived usefulness of the report and the presentation. For
these reasons, report preparation and presentation is of great importance.

10.2 QUESTIONNAIRE CHECKING

Figure 10.1 illustrates the report preparation and presentation process. The
process of report preparation is the final major step in a marketing research
project. It begins with the interpretation of the data analysis results pertaining
to the marketing research problem. The findings, supported by appropriate
conclusions and relevant recommendations, should be presented in a suitable
format to facilitate decision making by the client. Prior to writing the report,
discuss with the client the findings, conclusions and recommendations so that the
final product is helpful in meeting their needs. During these discussions, specific
dates should be confirmed for the delivery of the written report and other data as
requested by the sponsoring management.

The entire marketing research project should be summarised in one or more

written reports geared to the respective readers and an oral presentation may
also be made. After the client has had an opportunity to read the report, the
researcher should undertake necessary follow-up action. The researcher should
also assist the client in understanding the report, implementing the findings and
undertaking a thorough evaluation of the entire marketing research project.
 TOPIC 10 REPORT PREPARATION AND PRESENTATION  163

Problem Definition, Approach, Research Design, and Field W ork

Data Analysis

Interpretation, Conclusions and Recommendations

Report Preparation

Oral Presentation

Reading of the Report by the Client

Research Follow-up

Figure 10.1: The report preparation and presentation process

Source: Malhotra, N. (2002). Basic marketing research: Application
to contemporary issues. Upper Saddle River, New Jersey, USA: Prentice-Hall, Inc.

10.3 REPORT PREPARATION

What are the items that should be included in a research report?

Researchers differ in the way they prepare a research project. The personality,
background, expertise and responsibility of the researcher, along with the decision
maker to whom the report is addressed, interact to give each report a unique
character. There are guidelines for formatting and writing reports and designing
tables and graphs.
164  TOPIC 10 REPORT PREPARATION AND PRESENTATION

10.3.1 Report Format

Although the report formats are likely to vary with the researcher, the client and
the nature of the research project, a common framework can be designed for
writing the report. From this framework, the researcher can develop a specific
format tailored for a particular research project, after taking into account the
required level of formality and the degree of detail needed. Figure 10.2 summarises
the report parts.

Report Parts

Prefatory Parts Main Body Appended Parts

Figure 10.2: Report parts

Source: Zikmund, W. G. (2003). Business research methods, (7th ed.). South-Western,
Ohio, USA: Thompson Learning.

A typical report may contain the following items:

(a) Title Page

(b) Letter of Transmittal

(c) Letter of Authorisation

(d) Table of Contents

(i) List of tables
(ii) List of graphs
(iii) List of appendices
(iv) List of exhibits

(e) Executive Summary

(i) Major findings
(ii) Conclusions
(iii) Recommendations
 TOPIC 10 REPORT PREPARATION AND PRESENTATION  165

(f) Problem Definition

(i) Background to the problem
(ii) Statement of the problem

(g) Approach to the Problem

(h) Research Design

(i) Type of research design
(ii) Information needs
(iii) Data collection from secondary sources
(iv) Data collection from primary sources
(v) Scaling techniques
(vi) Questionnaire development and pretesting
(vii) Sampling techniques
(viii) Field work

(i) Data Analysis

(i) Methodology
(ii) Plan of data analysis

(j) Results

(k) Limitations and Caveats

(l) Conclusions and Recommendations

(m) Exhibits
(i) Questionnaire and forms
(ii) Statistical output
(iii) Lists

This format closely follows the earlier steps of the marketing research process.
The format should be flexible so that it can accommodate the unique features of
specific projects.
166  TOPIC 10 REPORT PREPARATION AND PRESENTATION

Figure 10.3 summarises the prefatory parts.

Prefatory Parts

Title page

Summary
Letter of
transmittal Objectives

Results
Letter of
authorisation
Conclusions

Recommendations
Table of contents

Figure 10.3: Prefatory parts

Source: Zikmund, W. G. (2003). Business research methods, (7th ed.). South-Western,
Ohio, USA: Thompson Learning.

(a) Title Page

The title page should include the title of the report, information about the
researcher or the organisation conducting the research, the name of the
client for whom the report was prepared and the date released. It should
also indicate the nature of the project.

(b) Letter of Transmittal

The letter of transmittal is a letter that delivers the report to the client and
summarises the researcherÊs overall experience with the project without
mentioning the findings.

(c) Letter of Authorisation

The client writes this letter to the researcher before work on the project
begins. It authorises the researcher to proceed with the project and specifies
its scope and the terms of the contract.
 TOPIC 10 REPORT PREPARATION AND PRESENTATION  167

(d) Table of Contents

This is based on the outline of the report and lists the sequence of topics
and their corresponding page numbers. In most reports, only the major
headings and subheadings are included in the table of contents. The lists
of the tables, graphs, appendices and exhibits should follow the table of
contents.

(e) Executive Summary

This summary describes the problem, approach and research design adopted
in the marketing research project. It is mainly devoted to the major results,
conclusions and recommendation.

(f) Problem Definition

It contains a statement of the management decision problem and the
marketing research problem. The problem definition also states the
background to the problem, highlighting the discussions with the decision
maker(s) and industry experts. It also mentions the secondary data analysis,
the preliminary research that was conducted and the factors that were
considered.

(g) Approach to the Problem

This section discusses the broad approach that was adopted in addressing
the problem. It also contains a description of the theoretical foundations
that guided the research, any analytical models formulated, research
questions and hypotheses.

(h) Research Design

This section details the nature of the research design adopted, the
information needs, the data collection methods, the scaling techniques,
questionnaire development and pre-testing, sampling techniques, and field
work. The material is presented in non-technical terms and the technical
details are included in the appendix. This section also justifies the
procedures adopted.

(i) Data Analysis

This section describes the plan of data analysis, justifying the selection
of the specific techniques. The data analysis techniques are described in
simple, non-technical terms.
168  TOPIC 10 REPORT PREPARATION AND PRESENTATION

(j) Results
This section which presents the findings of the research is normally the
longest part of the report and may comprise several topics. The results are
presented not only at the average level but also at the subgroup level.
Results must be geared directly in a coherent and logical way. The details
should be presented in tables and graphs, with the main findings discussed
in the text.

(k) Limitations and Caveats

All marketing research projects have limitations caused by time, budget
and other organisational constraints. The research design adopted may be
limited in terms of various types of errors that may be serious enough to
warrant discussion. This section must be written with great care and a
balanced perspective. On the other hand, this section should not erode the
readerÊs confidence in the research or unduly minimise its importance.

(l) Conclusions and Recommendation

This section contains the inferences derived from the marketing research
results and states the feasible and practical recommendations that can be
used by the client for its own decision-making purpose.

10.3.2 Report Writing

The following guidelines should be incorporated in a high-quality report:

(a) Reach out to the Reader

The report should reach out to the target audience. This can be done
by taking into account the clientÊs technical sophistication, interest in the
project, and the circumstances or manner in which they are going to read
and use the information.

(b) Ease of Comprehension

The report should be easy to follow. Clarity, brevity and logical structuring
of the presented material increase the readerÊs ability to comprehend the
report.

(c) Presentable and Professional Appearance

An appealing and inviting appearance created through skilful use of
typing, binding and reproduction can entice the reader and enhance a
desire to read the report.
 TOPIC 10 REPORT PREPARATION AND PRESENTATION  169

(d) Objectivity
The report should be truthful and sincere in its purpose. Any deviation
from this approach destroys the credibility of both the report and the
researcher.

(e) Reinforcement
Reinforcement of the information in the text through visual aids facilitates
communication with the reader and adds to the impact of the report.

(f) Terseness
Brevity in the presentation of a report, without compromising completeness,
adds to its appeal.

10.3.3 Guidelines for Tables

Statistical tables are a vital part of the report and deserve special attention. The
following are the guidelines for the table:

(a) Title and Number

Every table should have a number and a title. The title should be brief
and clearly describe the information provided. Arabic numbers are used to
identify tables so that they can be referred to easily in the text.

(b) Arrangement of Data Items

Data should be arranged in a table to emphasise the most significant
aspects of the data.

(c) Basis of Measurement

The basis or unit of measurement should be clearly stated.

(d) Leader, Rulings, Spacing

A leader is a dot or hyphen used to lead the eye horizontally, impart
uniformity and improve readability. Instead of ruling the tables horizontally
or vertically, use white spaces to set off data items. Skipping lines after
different sections of the data can also assist the eye in reading.

(e) Explanations and Comments: Heading, Stubs and Footnotes

Explanations and comments clarifying the table can be provided in the
form of captions, stubs and footnotes. A heading is a designation placed
over the vertical columns. Designations placed in the left-handed columns
are called stubs. Information that cannot be incorporated into the table
should be explained in footnotes.
170  TOPIC 10 REPORT PREPARATION AND PRESENTATION

(f) Sources of the Data

If the data contained in the table is secondary, the source of the data should
be cited.

10.3.4 Guidelines for Graphs

Generally, graphic aids should be used whenever practical. Graphical display of
information can effectively complement the text and tables to enhance clarification
of communication and impact. Types of graphical aids:

(a) Geographic and Other Maps

Used for communicating relative location and other comparative
information.

(b) Round or Pie Charts

Used for depicting proportions associated with a single variable. They are
unsuitable for displaying relationships over time or relationships among
several variables.

(c) Line Charts

Used for illustrating trends and changes over time, for comparing several
series of data and for interpolating and extrapolating forecast.

(d) Pictographs
This is the most creative type of graph and is useful for eye-catching
scenarios. However, pictographs do not depict results precisely.

(e) Histograms and Bar Charts

These charts are used to present absolute and relative magnitudes,
differences and changes. They are inappropriate for representing a series of
data plotted over time for more than one variable.

(f) Schematic Figures and Flowcharts

Used for displaying the steps involved in or the components of a process,
as well as for classification diagrams. They do not represent statistical data
analysis.
 TOPIC 10 REPORT PREPARATION AND PRESENTATION  171

10.3.5 Oral Presentation and Dissemination

The following are important aspects of making an effective oral presentation:

(a) Preparation and Rehearsal

The presenter(s) should be thoroughly familiar with the report in order to
explain it to management. Extensive rehearsal will result in a presentation
that is knowledgeable and polished and that covers all the important topics
in the time allotted by management.

(b) Presentation should be Geared to the Audience

The presentation should consider the audienceÊs background, interest and
involvement in the project as well as the degree to which they are likely to
be affected by it. This will make the presentation more meaningful to the
audience.

(c) Use of Visual Aids

Visual aids clarify and support arguments made in the presentation.

(d) Interact with the Audience

Sufficient audience interaction can be accomplished through good eye
contact, opportunities for questions and the use of stories, quotations and
anecdotes.

(e) Body Language

Descriptive gestures are used to clarify or enhance verbal communication.
Varying volume, pitch, voice quality, articulation and rate while speaking
emphasises key points of the oral presentation.

The key to an effective presentation is preparation. A written script or detailed

outline that follows the format of the written report should be prepared. The
presentation must be geared to the audience and should be rehearsed several
times before it is presented to management. Visual aids should be displayed
using a variety of media. It is important to maintain eye contact and interact with
the audience during the presentation.
172  TOPIC 10 REPORT PREPARATION AND PRESENTATION

Sufficient opportunity should be provided for questions. Appropriate stories

should be used to make the presentation interesting and convincing. Filler
words like „uh,‰ „you know‰ and „all right‰ should be avoided. Body language
is also important. It helps the speaker to convey the ideas in the report more
emphatically. Effective body language can reinforce the issue or the point the
speaker is trying to communicate to the audience. The speaker should vary
the volume, pitch, voice quality, articulation and rate while speaking. The
presentation should terminate with a strong closing. The „tellÊem‰ principle is
effective for structuring a presentation. This principle states:

(i) Tell them what you are going to tell them;

(ii) Tell them; and then

(iii) Tell them what you have told them.

Another useful guideline is the „KISS Âem‰ principle, which states: Keep It
Simple and Straightforward (hence the acronym KISS).

10.3.6 Research Follow-up

The researcherÊs task does not end with the oral presentation. Two other tasks
remain. First, the researcher should help the client understand and implement
the findings and take any follow-up action. Second, while it is still fresh in his or
her mind, the researcher should evaluate the entire marketing research project.

(a) Assisting the Client

(i) Clarify difficulties in understanding the report.

(ii) Assist in implementation of the findings.

(iii) Identify any other research projects that need to be undertaken.

(b) Evaluating the Research Project

(i) Identify the difficulties faced in doing the market research project.

(ii) Identify the shortcomings, if any, associated with the different stages
of the entire market research project.

(iii) Anticipate the possible impact if alternative methodology was

followed at different steps, such as data collection, data analysis,
report writing and presentation and so forth.
 TOPIC 10 REPORT PREPARATION AND PRESENTATION  173

EXERCISE 10.1

Tick „True‰ or „False‰ in the appropriate boxes.

No Statements True False

1. The first step in the report preparation and
presentation process is problem definition.
2. A letter that delivers the report to the client and
summarises the researcherÊs overall experience with
the project without mentioning the findings is called
the letter of authorisation.
3. The client writes a letter of estimation to the researcher
before the work of the project begins.
4. The executive summary is an extremely important part
of the report as this is often the only portion of the
report that executives read.
5. The problem definition section of the report should
include the nature of the research design adopted,
information needed, data collection from secondary
and primary sources, scaling techniques, questionnaire
development and pretesting, sampling techniques and
field work.
6. Data should be arranged in a table to emphasise its
most significant aspects.
7. If locating items is critical, an alphabetical arrangement
is the most appropriate.
8. An attractive way of illustrating trends and changes
over time is to use a line chart.
9. Bar charts can be used to display the steps or
components of a process.
10. Instead of ruling the table horizontally or vertically, it
is suggested that leaders be used to set off data items.
174  TOPIC 10 REPORT PREPARATION AND PRESENTATION

EXERCISE 10.2

Multiple Choice Questions

1. The first step in the report preparation and presentation process

is _____________________.
A. data analysis
B. problem definition
C. recommendation
D. research follow-up

2. The ___________________ should identify the need for further

action on the part of the client, such as implementation of the
findings or further research that should be undertaken.
A. letter of transmittal
B. letter of estimation
C. letter of authorisation
D. letter of recommendation

3. The__________________ is an extremely important part of the

report as this is often the only portion of the report that
executives read.
A. Executive summary
B. Problem definition section
C. Research design section
D. Conclusions and recommendations section

4. Which of the following is NOT a suggestion to follow when

writing a marketing research report?
A. Technical jargon should not be used.
B. The report should be easy to follow.
C. Objectivity should guide report writing.
D. Reinforce key information in the text with tables, graphs
and pictures.
 TOPIC 10 REPORT PREPARATION AND PRESENTATION  175

5. Instead of ruling the table horizontally or vertically, it is suggested

that __________________ be used to set off data items.
A. white spaces
B. leaders
C. stubs
D. hyphens

6. A __________________ is a chart that displays data in bars positioned

horizontally or vertically.
A. line chart
B. pictograph
C. bar chart
D. pie chart

7. The __________________ section of the report should include the

nature of the research design adopted, information needed, data
collection from secondary and primary sources, scaling techniques,
questionnaire development and pretesting, sampling techniques and
field work.
A. Executive summary
B. Problem definition
C. Research design
D. Conclusions and recommendations

8. Which of the following sections of the report gives the background to

the problem, highlights discussions with the decision makers and
industry experts and discusses the secondary data analysis, the
qualitative research that was conducted and the factors that were
considered?
A. Executive summary
B. Problem definition
C. Research design
D. Conclusions and recommendations
176  TOPIC 10 REPORT PREPARATION AND PRESENTATION

9. Data should be arranged in a table to emphasise the most

__________________.
A. expensive aspect of the data
B. current aspect of the data
C. significant aspect of the data
D. reliable aspect of the data

10. A letter that delivers the report to the client and summarises the
researcherÊs overall experience with the project without mentioning
the findings is called the __________________.
A. Letter of acceptance
B. Letter of authorisation
C. Letter of transmittal
D. Letter of recommendation

 Report preparation and presentation is the final step in the marketing

research project. This process begins with the interpretation of data analysis
results and leads to conclusions and recommendations.

 Next, the formal report is written and an oral presentation is made.

 After management has read the report, the researcher should conduct a
follow-up, assisting management and undertaking a thorough evaluation of
the marketing research project.

Appended parts Prefatory parts

Main body Report format
 MODULE FEEDBACK
MAKLUM BALAS MODUL

If you have any comment or feedback, you are welcome to:

1. E-mail your comment or feedback to [email protected]

OR

2. Fill in the Print Module online evaluation form available on myVLE.

Thank you.

Centre for Instructional Design and Technology

(Pusat Reka Bentuk Pengajaran dan Teknologi)
Tel No.: 03-27732578
Fax No.: 03-26978702