Search for notes by fellow students, in your own course and all over the country.
Browse our notes for titles which look like what you need, you can preview any of the notes via a sample of the contents. After you're happy these are the notes you're after simply pop them into your shopping cart.
Title: MARKETING RESEARCH
Description: Student will gain a deeper understanding of marketing research and its value in analyzing consumer markets and the environment.
Description: Student will gain a deeper understanding of marketing research and its value in analyzing consumer markets and the environment.
Document Preview
Extracts from the notes are below, to see the PDF you'll receive please use the links above
KENYATTA UNIVERSITY
DIGITAL SCHOOL OF VIRTUAL AND OPEN LEARNING
IN COLLABORATION WITH
SCHOOL OF BUSINESS
DEPARTMENT: BUSINESS ADMINISTRATION
UNIT CODE & NAME: BBA 843 – MARKETING RESEARCH
WRITTEN BY:
DR
...
Topics include an overview of market research and research
design, exploratory research; descriptive research; scaling; sampling; and data analysis and
reporting
...
Clearly articulate the differences in conducting various research designs to define the
research problem
...
Create and conduct a small survey, applying a wide range of survey, scale, and
questionnaire techniques
...
3
TABLE OF CONTENTS
Contents
1
...
Market Research Concept
...
1
...
9
1
...
Meaning of Research
...
3
...
11
1
...
Role of Marketing Research in Decision Making
...
5
...
14
1
...
1
...
15
1
...
Marketing research and Marketing Information System
...
7
...
17
1
...
1
...
17
1
...
2
...
17
1
...
3
...
18
1
...
4
...
18
2
...
Market Research process
...
1
...
19
2
...
1
...
2
...
20
Summary
...
2
...
Note
...
2
...
Activities
...
2
...
Further Reading/ References
...
2
...
Self-Test Questions
...
0
...
29
3
...
Lecture Objectives
...
2
...
30
3
...
Considerations in identifying marketing problems
...
4
...
31
3
...
Marketing Research Objectives
...
7
...
34
3
...
Summary
...
8
...
Note
...
8
...
Activities
...
8
...
Further Reading/ References
...
8
...
Self-Test Questions
...
0
...
37
4
...
Lecture Objectives
...
2
...
38
4
...
1 Exploratory research design
...
2
...
Descriptive research designs
...
2
...
Causal/ Experimental research Design
...
2
...
Mixed methods/Action research design
...
3
...
50
4
...
1
...
50
4
...
2
...
50
4
...
3
...
51
5
4
...
4
...
51
5
...
Sample Selection
...
1
...
52
5
...
Unit of analysis
...
2
...
Population
...
2
...
Sampling
...
2
...
The Sampling Process
...
2
...
Sampling Designs:
...
2
...
Sample Size determination
...
3
...
60
5
...
1
...
60
5
...
2
...
60
5
...
3
...
61
5
...
4
...
61
6
...
Data Collection
...
1
...
62
6
...
Types of Data
...
2
...
Secondary Data Sources
...
2
...
Primary Data Sources
...
3
...
67
6
...
1
...
67
6
...
2
...
72
6
...
3
...
73
6
...
4
...
75
6
...
5
...
76
6
6
...
6
...
4
...
Document Analysis
...
77
6
...
1
...
77
6
...
2
...
77
6
...
3
...
78
6
...
4
...
78
Tests of Sound measurement and levels of scale
...
1
...
79
7
...
Concept of measurement in research
...
2
...
Measurement process
...
3
...
81
7
...
1
...
4
...
87
7
...
1
...
87
7
...
2
...
89
7
...
3
...
92
7
...
8
...
83
Summary
...
5
...
Note
...
5
...
Activities
...
5
...
Further reading
...
96
8
...
Objectives
...
2
...
97
8
...
1
...
3
...
97
Data Analysis for Qualitative Research
...
4
...
100
8
...
1
...
5
...
103
8
...
1
...
104
8
...
2
...
106
8
...
3
...
106
8
...
9
...
101
Summary
...
6
...
Note
...
6
...
Activities
...
6
...
Further reading
...
6
...
Self-Test Questions
...
111
9
...
Objectives
...
2
...
111
9
...
Types of Reports
...
4
...
112
9
...
Parts of the research report
...
6
...
117
9
...
Summary
...
7
...
Note
...
7
...
Activities
...
7
...
Further reading
...
7
...
Self-Test Questions
...
0
...
It is therefore very paramount for
marketing managers to understand the nature of marketing research and to able to evaluate
the kind of information to seek, how to get it, and how to utilize it for effective marketing
decisions
...
1
...
Objectives
By the end of the lecture, you should be able to:
Explain the meaning of research
Discuss the various types of research
Identify the role of marketing research in decision making
Discuss the features of marketing research
Evaluate factors influencing the decision to do market research
9
1
...
Meaning of Research
Marketing research is a terminology made up of two components: Marketing and research
...
American Marketing Association- Marketing as the process of all activities necessary for the
conception, pricing promotion, and distribution of ideas, goods, and services to create
exchanges tat satisfies individual and organizational objectives
...
Market concept philosophy holds that the KEY to achieving organizational goals consist of
being more effective than competitors in integrating marketing activities toward determining
and satisfying the needs and wants of target markets
...
Marketing research- This is the process of systematically analyzing and interpreting data
pertaining to the company’s market, customers, and competitors with a goal of improving
marketing decisions
...
European society for opinion and marketing research ( ESOMAR) defines market research
as:
“A key element within the field of marketing information
...
Generate, refine, and evaluate marketing actions;
improve understanding of marketing as a process and of the ways in which specific
marketing activities can be made more effective
...
Objective-methods used should eliminate bias, error and should not be subjective
...
It as such links the customer,
consumer and public with the marketer through information that is used to identify and
define marketing opportunities and problems
...
It specifies the information required to address issues, designs
the methods for collecting information and manages and implements data collection,
analyzes results and communicates the findings and their implications
...
g
...
A specific marketing research process on the contrast is
intended to meet nonroutine information needs that cannot be met adequately by a firm’s
MIS
...
3
...
Applied Research - (marketing research primarily conducted by practitioners, or for them by
academicians) is research aimed at solving a specific, pragmatic problem such as for better
understanding of the marketplace, determination of why a strategy or tactic failed, and
reduction of uncertainty in management decision making
...
In most of its forms
it does this by using a cyclic or spiral process that alternates between action and critical
reflection and in the later cycles, continuously refining methods, data and interpretation in
the light of the understanding developed in the earlier cycles
...
11
Action research therefore has some components which resemble consultancy or change
agency, and some which resemble field research
...
Similar to all other types of research, research methodologies can be classified into two broad
categories:
Quantitative Research
Qualitative Research
Quantitative research refers to an inquiry method/approach useful for describing trends and
explaining the relationships among variables
...
Data collected is usually objective and
minimizes the extent of researcher bias
...
4
...
Descriptive function- it has the role of identifying trends in the marketing field so as
to identify opportunities and problems in the market place
...
2
...
This is done through break-even analysis, costbenefit analysis, market-mix evaluation testing, and concept testing
...
Predictive function- It is the role of market research to enable the researchers to
evaluate cause and effects among various variables
...
Such as what amount of promotional budget
may result to a predicted amount of sales
...
It is specifically used to track and monitor product sales, customer satisfaction trends
etc
...
5
...
Relevant Research
Research should not be conducted to satisfy curiosity or confirm the wisdom of previous
decisions
...
This information is the backbone of the
ongoing information system
...
2
...
Often these decisions are fixed in
time and must be taken according to a specified schedule using whatever information is
available
...
In cases where the timing of decisions is contingent upon research
results, its important to note that he failure to take corrective action or pursue an opportunity
as quickly as possible will result in opportunity costs
...
Efficient Research
There are two senses in which marketing research must be efficient:
a) The first aims at reducing expenditure on research-get maximum quality of information
with minimum expenditure of time and money
...
The study should be expensive and elaborate only if the decision is important and the
research information will be helpful and timely
...
Accurate research
Despite careful research designs, it is inevitable that biases arise due to question wording or
interpretation, the sampling plan and other elements of the research design
...
If several
approaches with different kinds of biases yield similar conclusions, the accuracy will be
enhanced
...
Ethical research
Ethical conduct in research concerns the rights and responsibilities of four parties namely:
society, the research client, the researcher and the respondent
...
It is
expected to be aware of research findings that give implications on the health and well being
of the population
...
14
Cont: 5 Ethical Research
• Client rights-Clients have a right to confidentiality of the working relationship with the
research agency
...
Clients have a right to expect quality research from the
research agency including protection from unqualified researchers or untruthful presentation
of research findings
...
g
...
Similarly, researchers also expect
specific issues from respondents e
...
not to deliberately distort or conceal data, and to respect
the confidentiality of the data so requested
...
1
...
1
...
Time constraint- the urgency of a situation precludes the use of research
...
Availability of data-research is needed when managers lack adequate information to make
sound decisions
...
Nature of decision-if it is of considerable strategic or tactical importance
...
Benefit versus cost-consider whether the value of the research information exceeds the
cost of conducting the research
...
• When decision-making information already exists
...
15
1
...
Marketing research and Marketing Information System
Marketing information system is a planned system of collecting, processing, storing, and
disseminating data in the form of information needed to carry out the functions of
management
...
The source of information include: internal data,
marketing intelligence, and marketing research
...
Marketing intelligence- is the systematic collection and analysis of publicly available
information about competitors and trends in the marketing environment
...
Marketing research- the systematic design, collection, analysis, and reporting data
relevant to specific marketing situations facing an organization
...
E
...
break-even analysis, regression models etc
...
7
...
There are various categorizations of research
...
There is a strong link between marketing research and marketing management, since research is
done to aid decision-making
...
7
...
Note
You should note that, Marketing research is an important component of the marketing
information systems
...
7
...
Activities
Consider all the products and/or services offered by your organization (or any organization of
your choice)
...
17
1
...
3
...
Aaker, D
...
and Day, S
...
(1990) Marketing Research 4th Ed
...
2
...
T
...
Butterworth-hiennemann Oxford
Blankenship, A
...
Breen, G
...
and Dukta, A
...
Contemporary publishing group Inc
...
3
...
W
...
and Stasch S
...
7th Ed
...
USA
...
Baker, M
...
(1991) Research for marketing, The Macmillan Press Ltd Hong Kong
...
7
...
Self-test questions
Compare and contrast the quantitative and qualitative research methodologies
Why is it critical for marketing managers to adhere to ethical considerations?
Evaluate the role of marketing research in marketing decision making?
18
2
...
Market Research process
Market research is a procedural process of identifying a research problem, and identifying feasible
ways of collecting the required information,in an objective manner in order to make marketing
management decision
...
It explains how researchers can translate management problems to research
problems
...
1
...
1
...
Research Process
The marketing research process is a set of steps that define the tasks to be accomplished in
conducting a marketing research study
...
Problem / opportunity identification
Generally a marketing problem or opportunity exists when there is a gap or difference
between actual condition or state and a condition or state desired by an organization
...
If only one action is available (or none at all), or if there is certainty
about the outcomes of the alternative, there is no problem
...
Opportunities are situations with positive potentials, which if recognized and capitalized on,
are profitable to the enterprise
...
b) Difficulties are developments that have negative or counteractive consequences
...
The formal quantitative research should not begin until the problem has been clearly defined
...
Problem audit
This is a comprehensive and systematic examination of a marketing problem with the
objective of understanding its origin and nature
...
An audit helps distinguish between problem symptoms and problem causes e
...
a company
experiencing sales decline for its infant cereals conducted an audit which revealed that the
cause of the decline was the failure to use the scheduled retail outlets
...
There is the need for proper communication and
cooperation between the marketing managers and the marketing researcher i
...
the researcher
should understand the context of the marketing problem and terminology used in describing
it
...
Audits should also be used
to help marketing managers to identify those marketing problems that are impractical to
research on e
...
it is difficult to study the repeat purchase behaviour for products with life
expectancy of 8 years (repeat purchases are minimal), or characteristics of people who falsely
redeem newspaper coupons (such people wouldn’t wish to be identified)
...
Its goal is to familiarize the
marketing researcher with the environment of the problem
...
Several activities are undertaken in this
phase:
Situation investigation-looks at how the problem started, why, the purpose of the research,
potential practical and political consequences of the research
...
Includes
looking at public material and public documents, surfing the Internet and online databases
...
Case studies-investigate a small number of cases i
...
people, organizations, and situation
from a clinical perspective
...
g
...
A situation analysis will give information on the nature of target market, the marketing mix,
competition and other external factors
...
The researcher decides on what information he needs to answer the
remaining questions
...
Step 2
...
The research objectives explain the purpose of the research in measurable terms
...
Objectives consider how the information provided by the research is to be
used for decision-making, what form it will take, and to whom the information is reported
...
They contain the particular variables to be investigated
...
The objective would be stated:
To determine awareness level of potential buyers of proposed office furniture with respect
to the furniture brand currently offered
21
Step 3: Marketing research Project Design
This systematically sets forth what research data are to be collected as well as how the data
are to be collected and analyzed
...
It details the
procedures necessary for obtaining the required information, and its purpose is to design a
study that will test the hypothesis of interest, determine the possible answers to research
questions, and provide the information needed in decision making
...
A marketing
research design can be categorized as either quantitative or qualitative
...
Whereas the qualitative research designs include exploratory, narrative, grounded research,
historical, ethnographic and they involve an in-depth analysis of a phenomenon which is
collected through non-structured means such as observation or interviews
...
Quantitative research designs involve the scientific description of a social phenomenon in a
mathematical or computational technique using numbers
...
The researcher has to carefully plan
the study including the tests or data collection instruments, each subject is identically studied
with very little room for human bias
...
These designs indicate that research is a
work in progress that may change as the data are collected
...
The data collection instrument used are flexible
and non-structured, data is interpreted according to researchers experiences and intuition
...
A researcher may decide to include everyone in the population hence
conducting a census, however, due to the constraints such as time and cost limitations,
researchers opt to select a representative group from the entire population also known as the
sample, to derive inferences about the population through a keen observation of the sample
...
There are three types of population: universal, target and
accessible populations
...
Depending on the characteristics of the population, researchers select the samples through
probability or non-probability means
...
Whereas, situations where the size or the number of people in
the population are not known, and hence not everyone has a chance of being selected in the
sample, in such cases, the non-probability sampling techniques are used
...
Probability Sample - A chosen subset of the population of interest that ensures a
representative cross section by giving every element in the population a known, nonzero
chance of being selected
...
2
...
This is done by selecting the sample
from the population in a manner that is not random
...
The researcher has no control over this error, however it can be minimized by increasing the
sample size so that it can accurately reflect the population
...
Sampling bias- This is the error that is in control of the researcher
...
Researchers commit this error
when they use inappropriate sampling techniques to select their respondents, e
...
using nonprobability sampling techniques such as convenience sampling to select a samples from a
23
population with finite number of people
...
If the research
design is of quantitative nature, then a structured closed-ended questionnaires are used to
collect the data whereas, if the research design is qualitative in nature, then a non-structured
flexible technique such as observations, focus, groups, interviewing, or open ended
questionnaire is used
...
There are two types of data sets: primary data and secondary
data
...
It is collected first hand to respond to the study at hand
Secondary data is data collected at another time prior to the timing of the current study and
that was collected for a different research purpose
...
However, various considerations need to be kept in mind when deciding on the
prototype of the data to use
...
The data collection stage also involves the choice of data collection instruments, mechanisms
of testing if they are sound measurement tools and identifying the measurements of scales to
use when developing the instruments
...
The type of tool to be used depends on the design of the study
...
Researchers select the instrument that best fits their research methodology
and that enables them to respond to the research questions
...
24
Step 6: Data Analysis and Interpretation
Once the data has been collected, the researcher perfoms several procedures to convert the
raw data into a format that will enable in responding to the pertinent research questions
...
Hence, data analysis
refers to examining what has been collected in a survey or experiment in order to make
deductions or inferences
...
The data is
summarized by evaluating its relevance to the research problem,
Data Processing- this involves the process of editing/pre-processing, coding, classifying, and
finally tabulation
...
This can be electronically or non-electronically, there are a variety of
software’s that have been developed to store data and enhance its analysis such as Excel
spreadsheets, SPSS, among others
...
The data can be
divided into discrete data or continuous data
...
Qualitative research involves intensive data collection over an extended period of time in
natural setting
...
However, it is important to note, the kind of data collection instrument used is what
determines if the data is quantitative or qualitative in nature
...
Qualitative data is analysed using thematic analysis which refer to
topical or subject discussions or content analysis which is a systematic description of the
form or content of written or spoken material
...
The entire project is documented in a written report that addresses the specific research
questions identified; describes the approach, research design, data collection and data
analysis procedures used; and presents the results and major findings
...
Managers to make decisions then
use this report
...
That is, if findings don’t guide decisions, the research is a
waste of time and money
...
The lack of
boundaries by researchers may result to resource mis-use, it could result to the research not
accomplishing its goals, in cases where the researcher has not created appropriate rapport
with the respondents and it could also have legal implications in cases where the study has
been conducted without obtaining appropriate or adequate permission or the study has issues
of plagiarism that undermine the intergrity of the researcher
...
2
...
Summary
The marketing research process is therefore a systematic, cyclical, and logical process
...
26
2
...
1
...
A well orchestrated research has all the sections working in tandem to each other
...
2
...
Activities
Using the product that you identified in the first lecture, develop an outline of the research
proposal identifying the key issues to address in each of the sub-sections
...
2
...
Further Reading/ References
5
...
A
...
G
...
John Wiley and
sons, New York
...
Quee, W
...
(1999) Marketing research 3rd Ed
...
B
...
E
...
(1998) State of the art Marketing r
Research 2nd ed
...
USA
...
Boyd, H
...
Westfall, R
...
F (1990) Marketing Research texts and
cases
...
Richard D Irwin Inc
...
8
...
J
...
27
2
...
4
...
Why is it important for researchers to be mindful about the choice of research
objectives in a study
2
...
Discuss ways in which as a researcher you can minimize the extent of sampling
errors and bias in your study
28
3
...
Defining a Market Research Problem
A marketing research problem should always be defined in a manner that will allow the
researcher to obtain all the information needed to address the management decision problem,
and also guide the researcher in proceeding with the project
...
1
...
2
...
Only when the marketing research problem has been clearly defined can
research be designed and conducted properly
...
The tasks involved include discussions with the key decision makers including a problem
audit, interviews with industry experts, analysis of secondary data and qualitative research
...
This kind of analysis helps in the identification of a management
decision problem, which should then be translated into a research problem
...
If the management decision
problem is should the advertising campaign be changed? Then this can be translated as: To
determine the effectiveness of the current advertising campaign
...
To
deliver products that solve target market customer needs a researcher must first identify
market problems
...
Stated
needs are those problems that the customers have already expressed explicit about what the
marketers or organizations ought to address
...
For such stated needs, banks can evaluate
the effect of service automation on customer satisfaction, whereas stated needs are important
and every manager must ensure that they are met, they are not as powerful as the silent
needs
...
A Kenyan example would be the M-Pesa service
...
30
3
...
Considerations in identifying marketing problems
The following are the considerations that marketing managers should bear in mind when
identifying the marketing problems:
-
Why do this market research ( is it feasible, cost- benefit effects, etc)
-
What action will be taken once the research is complete?
-
What has caused the problem or led to the opportunity?
-
What is already known about the research problem?
-
What target population will be relevant in providing the key information?
-
What specific information is needed for the research: (market size, trends, buying
behavior, customer needs or segmentation)
-
What is the proposed budget
-
Are there any reporting requirements?
-
When are the findings required?
3
...
Identifying marketing problems
When identifying the marketing problems to investigate on
...
Innovation and competition- investigating on problems that focus on providing
innovative solutions and competitive strategies in a manner that provides greatest
customer satisfaction
...
focus on the customers- Marketers’ can do informal surveys to establish the unmet needs
from the customers
...
Reviewing the feedback channels of customers to the organization can provide areas or
gaps that can be filled through doing research
...
Focus on revenues- All businesses exist to make a profit
...
Hence, marketers can focus on research problems that enable the firm to be able to
establish reasons as to the drop in profitability among some of the products and also to
evaluate a decrease in the market share in any of the products
...
5
...
Whereas
marketing problems are wide, marketing research problems are narrow and focused
...
Examples:
Marketing management Problem
Marketing Research Problems
Develop a package for a new product
Evaluate effectiveness of new package
designs
Increase market penetration through
agencies in different locations
Evaluate prospective agency outlets
Increase amount of repeat purchasing
Assess the current amount of repeat
purchasing behaviors
32
3
...
Marketing Research Objectives
Every research should have defined and explicit research objectives which
clearly state why the research is being done
...
Meaning, all aspects of the research design, sampling
choice, target respondents, data collection instruments, and questionnaire are determined by the
choice of objectives
...
The objective of the
research could range from helping a company improve its satisfaction ratings among consumers,
to finding new markets for its products
...
For instance, if a company is struggling with stagnant sales ( marketing management problems) ;
the marketing research problem would be to establish products lines that have stagnant sales;
from which, various research objectives can be formulated as relates to the various products lines
to seek for reasons for the stagnant sales among them
...
These aspects could include:
price of the products, promotional techniques adopted by each of the product lines, the market
segment of the specific product lines, the brand image of the various product lines etc
...
To write out clear objectives, a marketer needs information about the background of a
problem
...
33
3
...
Market research Hypothesis
A research hypothesis is a statement about the expected relationship between observable or
measurable events or expected differences between groups under investigation
...
It is a
tentative answer regarding what the researcher considers as ought to be the possible outcome of
an existing problem
...
Hypothesis testing is used when researchers want to evaluate
the statistical effect of quantifiable aspects of their research
...
There are two main types of hypothesis: null and alternate hypothesis
...
It does not show a bias towards a
certain variable having an effect on the dependent variable
...
Alternate hypothesis on the other hand, refers to the researchers hypothesis
...
Note, in research, the role of research is to
validate the null hypothesis and not the alternate hypothesis
...
g
...
It should be reasonable and consistent with the existing body of knowledge
Should be testable and verifiable
Should be clear and brief
Should give the research the purpose by providing the direction of the study- hence should be
aligned to the research objectives
34
3
...
Summary
Formulation of the marketing research problem, marketing research objectives and
hypothesis sets the direction of the entire proposal
...
3
...
1
...
Market
hypothesis are only relevant for quantitative research problems
...
8
...
Activities
Identify a marketing management problem, from that problem, specify the market research
problem and derive specific research objectives and where possible research hypothesis to
evaluate the problem
35
3
...
3
...
Aaker, D
...
and Day, S
...
(1990) Marketing Research 4th Ed
...
10
...
T
...
Butterworth-hiennemann Oxford
Blankenship, A
...
Breen, G
...
and Dukta, A
...
Contemporary publishing group Inc
...
11
...
W
...
and Stasch S
...
7th Ed
...
USA
...
Baker, M
...
(1991) Research for marketing, The Macmillan Press Ltd Hong Kong
...
8
...
Self-Test Questions
Compare and contrast a marketing management problem from a market research
problem
Evaluate some silent needs in Kenya that innovative companies can evaluate and provide
solutions to
What is the role of research hypothesis in a study?
36
4
...
Research Design
This topic discusses research designs as they apply in marketing research
...
The
correct design will save time and money and will result in valid and reliable information
...
4
...
Lecture Objectives
By the end of the lecture, you should be able to:
Define research design and classify the various research designs
Explain the steps involved in various research designs
Compare and contrast quantitative and qualitative research designs
Develop a research problem specifying the research design to use
37
4
...
Research Design
A research design is a framework or blueprint for conducting a marketing research project
...
It specifies the methods and procedures for
conducting a specific research project i
...
it guides the implementation of a research study
towards the realization of the research objectives
...
A research design serves as a bridge between what has been established (the research
objectives) and what is to be done, in the conduct of the study, to realize those objectives
...
The main purpose of the design is that:
It forms the essential framework for research action
...
It ensures data collected meets research objectives
Fulfills information needs of the decision maker
...
The best research design is the one that offers the greatest value for money, or the one that
best allocates the available research funds
...
Based on these two classifications, research designs can serve the following
functions:
-
Exploratory- this is a qualitative research, where the researcher has not preassumptions of what the problem is and how it can be solved hence it takes the
attitude of exploring the general concept to crystallize a marketing problem or to
discover general ideas or insights that relate to the subject under study
...
Meaning, the data is
collected as observed without any manipulations by the researcher
...
-
Causal/ Experimental research designs- These are also quantitative research designs
and are referred to as the intervention research
...
They aim at explaining whether the intervention influences the
outcome for the group as opposed to another or if there are significant relationships
between the variables
...
These designs include the evaluation or testing of research
hypothesis
...
These research designs aim at establishing not only inferences or describing trends
within the population but also in understanding and exploring the in-depth aspects of
the research problem
...
2
...
It gives background knowledge about the problem area or product
field in question, and is qualitative in nature
...
This design often produces findings in the form of hypothesis for subsequent
verification through other forms of research
...
Structured data collection
forms are rare, and controlled experiments and random sampling are not compulsory
...
As an example, consider the problem of developing an identity for a new brand of women's
fragrance
...
It provides a
major cue to the consumer as to whether the product will be congruent with the consumer's
self-image
...
Both brand names turned out to have creative
personalities, but in different ways
...
The results were very useful in helping the marketing team communicate
with the creative developing the advertising
...
2
...
1
...
It gathers both internal and external data
...
External data can be gathered from government statistical yearbooks, bulletins, journals, or
trade association reports
...
Experience surveys/ expert opinions: This is applied in defining the marketing problem
where the researcher organizes informal talks with a group of resource people who can offer
a wealth of information
...
The purpose is to help formulate the problem and clarify concepts, hence
probability sampling is not required, but purposeful sampling to eliminate inarticulate people
...
Ethnographic survey: This is in situation where the inquirer interested in peoples’ behavior
chooses to explore the behavior of the target group in the natural settings
...
Focus group- This is applied in defining the marketing problem where the researcher
organizes informal talks with a group of resource people who can offer a wealth of
information
...
41
The purpose is to help formulate the problem and clarify concepts, hence probability
sampling is not required, but purposeful sampling to eliminate inarticulate people
...
Focus groups are best for studies that aim at new product development, positioning,
targeting, and segmenting strategies, habits and usage studies, packaging assessments,
attitude studies, advertising evaluations, promotional techniques evaluations, and idea
generation
...
In-depth interviews- This is a special form of focus group discussion in which one, and only
one participant takes part in the discussion session with the interviewer
...
These
unstructured one-on-one interviews are intended to discover deep-seated motivations i
...
oneon-one interviews that probe and elicit detailed answers to questions, often using clinical
nondirective techniques to uncover hidden motivations
...
It is preferred over focus group discussion when the research topic involves intimate
subject matter e
...
contraception, or very personal views e
...
attitudes towards management,
political opinions; or when the subject matter is potentially embarrassing e
...
sanitary towel
protection for women
...
g
...
Projective techniques- Ways of tapping respondents' deepest feelings by having them
"project" their own feelings onto unstructured situations
...
42
Types of projective techniques include:
1
...
This
projective technique is practical and effective, and often used to select brand names,
advertising campaign themes, and slogans
...
Sentence and Story Completion - Tests in which the respondents complete sentences or
stories (i
...
, structured scenarios) in their own words
...
They can be used in conjunction with word association tests, and they are most reliable
...
“
3
...
They are very versatile, and can be used to determine differential attitudes toward
two types of establishments and the congruity, or lack of congruity, between those
establishments and a particular product
...
4
...
," or
"most people
...
Often used to probe feelings about sensitive subjects that might be embarrassing or evoke
hostility if answered directly by a respondent
...
Thematic Apperception Test- “ Presents a series of pictures to research subjects and asks
them to provide a description or a story about the picture
...
Case studies allow existing information to be
reused in new ways that the original collectors of the data did not envisage
...
For example, a shirt manufacturer interested in
surveying retailers may first look at a few retail stores to identify the nature of any problems
or topics that a larger study should investigate
...
The researcher has no standard procedure to
follow, but must be flexible and attempt to glean information and insights wherever they
appear
...
But as
with other exploratory research, the results of case studies are only tentative but not
conclusive
...
The ultimate goal is to discuss all aspects of the groups'
ideas to reach a consensus on the rank order of their importance to the research problem
...
Since NGS is available as a superior replacement
for the traditional Focus Groups (FGs), you may be wondering why anyone would use Focus
Groups
...
Second, the moderator likely would bias the results
unintentionally unless s/he had the equivalent of a Masters degree in psychology because
preparation for and conducting the sessions hinges on effectively using clinical nondirective
techniques (like Depth Interviews discussed below) to uncover hidden motivations
...
Case studies allow existing information to be
reused in new ways that the original collectors of the data did not envisage
...
For example, a shirt manufacturer interested in
surveying retailers may first look at a few retail stores to identify the nature of any problems
or topics that a larger study should investigate
...
The researcher has no standard procedure to
follow, but must be flexible and attempt to glean information and insights wherever they
appear
...
But as
with other exploratory research, the results of case studies are only tentative but not
conclusive
...
The ultimate goal is to discuss all aspects of the groups'
ideas to reach a consensus on the rank order of their importance to the research problem
...
Since NGS is available as a superior replacement
for the traditional Focus Groups (FGs), you may be wondering why anyone would use Focus
Groups
...
Second, the moderator likely would bias the results
unintentionally unless s/he had the equivalent of a Masters degree in psychology because
preparation for and conducting the sessions hinges on effectively using clinical nondirective
techniques (like Depth Interviews discussed below) to uncover hidden motivations
...
2
...
Descriptive research designs
Descriptive research allows us to establish the frequency of occurrence of individual
variables, and the covariation among pairs of variables
...
Descriptive research aims at describing the characteristics of the population
under study
...
Data for this design is collected through personal interviews, mail interviews
or telephone interviews
...
Strict sampling procedures are essential to minimize systematic errors and to
maximize data reliability
...
This type of research design would for instance be appropriate when the research objectives
include one or more of the following:
a
...
b
...
c
...
d
...
Other examples of descriptive research include the following situations:
Market studies describe the market size, buying power of customers, availability of
distributors and consumer profiles
...
Sales analysis studies describe sales by geographic region, product line, and type of
the account and size of the account
...
Product usage studies describe consumption patterns
...
Pricing studies describe the range and frequency of price changes and probable
consumer response to proposed price changes
...
46
4
...
2
...
Types of Descriptive research Design
Descriptive research can involve either of two different forms namely, longitudinal studies
and cross sectional studies
...
Longitudinal studies
It is also termed as time series studies
...
For example trend data that relates
what and how market changes occur
...
In such a
case, the sampled groups need to remain identical if the comparisons are to be reliable
...
Cross sectional studies
One observation or measure is taken on one or more objects at a single point in time
...
Cross tabulations of data play a crucial role in
cross sectional studies
...
2
...
Causal/ Experimental research Design
The primary goal of causal research is to explore and establish a cause-and-effect relationship,
if any, between variables
...
The underlying principle is that when a single independent variable is manipulated while all
other variables are controlled over a period of time, any change in the dependent variable will
be attributable entirely to the independent variable
...
For example:
47
a) Will a 5% increase in the price of a product reduce the amount of the product a person will
buy?
b) Will redesigning the cereal box so it is shorter and less likely to tip over improve consumer
attitudes toward the product?
Causal research designs involve experimentation
...
An experiment measures the change in
the dependent variable created by a specific, controlled change in another variable(s), which
are called the independent variable(s)
...
Thus, the researcher is an active participant in the research process instead of a passive
collector of data as with the survey or observation methods of research
...
(This is
better described as "Concomitant Variation" when talking about causation studies
...
Correlation Refers to variables that vary together in a predictable manner
...
Are the cause and effect associated with each other? Associated
variation can be established using tests of statistical significance
...
2
...
e
...
That is, the cause must
precede the effect
...
3
...
48
4
...
3
...
Experimental Setting
These are of two types:
Laboratory Experiments –
In Laboratory experiments, the researcher creates situations that simulate the real conditions
e
...
pre-testing a TV commercial in a theatre or studio
...
However, while the laboratory allows the researcher to control the variables involved, the lab
may not accurately represent the real marketplace
...
Thus, lab results are said to have
good internal validity, but often lack external validity
...
Field Experiments –
Field experiments involve real life situations and are realistic, also known as market tests
...
This solves the problem of
realism of the test environment, but factors other than the independent variable(s) of interest
may influence the observed changes in the dependent variable of interest because the
researcher cannot control all other independent variables that may affect the dependent
variable
...
Thus, field experiments often lack internal
validity, while having better external validity
...
4
...
4
...
In mixed methods research, the researcher combines both
forms of data to best explain and explore the research problem
...
Conducting this type of study, requires that the researcher decided on the priority
or weight of each form of data as well as the sequence of data collection
...
49
4
...
Summary
The marketing research design section is a very critical section for every
research as it determines the procedures in which the data will be collected, from whom the
data will be collected, how it will be collected and finally how it will be analyzed and
interpreted
...
4
...
1
...
Researchers do not begin by identifying the research design to be used before developing the
research objectives or problem statement
...
3
...
Activities
Identify a marketing management problem, from that problem, specify the market research
problem and derive specific research objectives and where possible research hypothesis to
evaluate the problem
50
4
...
3
...
A
...
G
...
John Wiley and sons,
New York
...
T
...
Butterworth-hiennemann Oxford
Blankenship, A
...
Breen, G
...
and Dukta, A
...
Contemporary publishing group Inc
...
Boyd, H
...
Westfall, R
...
F (1990) Marketing Research texts and cases
...
Richard D Irwin Inc
...
Baker, M
...
(1991) Research for marketing, The Macmillan Press Ltd Hong Kong
...
3
...
Self- Test Questions
Compare and contrast quantitative and qualitative research designs
Discuss situations when it is most appropriate to conduct experimental research designs
What do we mean by Expo Facto research designs?
Researchers argue that in marketing research, qualitative research designs are more
appropriate than the quantitative designs explain providing examples
...
0
...
Since, research is a tedious, costly process, it is important that researchers identify
ways of selecting representative samples from the population so as to make generalizations
about the populations from the selected samples
...
In instances, where the population is finite,
meaning that the number of people making up the population is known, the researcher can use
probabilistic sampling techniques that provide everyone a chance of being selected unlike in
infinite populations in which members of the population are not known
...
5
...
Objectives
By the end of the lecture, you should be able to:
Compare and contrast the probability and non-probability sampling categories
Discuss the various methods/techniques under each of the categories: probability and nonprobability categories
Explain the key steps in sampling procedures
Discuss errors associated with sampling and how they can be mitigated
Conduct a sampling procedure for probability and non-probability sampling techniques
5
...
Unit of analysis
In identifying the choice of respondents, every researcher must identify the
unit of analysis
...
It is the
52
“what”, or “who” is being studied
...
Unit of
analysis refers to the level at which you pitch the conclusions
...
It is also
referred to as the target population
...
It is the aggregate that
conforms to a given specification
...
The population that is targeted by the
researcher is referred to as the “target population”
...
2
...
Population
Population refers to an entire group of individuals, events or objects having a common
observable characteristic
...
Universal population is everyone in the universe with
common observable characteristics
...
However, since it is not possible or even feasible to target everyone in the
entire universe, researchers mostly focus on the target population
...
Accessible population, is
similar to target population, however is limited to the geographical location accessible to
the researcher
...
The greater the variance between the sample and the population, the higher
the chance of sampling errors, hence becoming difficult to establish population validity
...
5
...
2
...
A sample is a smal group of individuals or objects
selected from the population in such a manner that it represents population characteristics
...
However, there are
limitations that occur in sampling: chance factor, bias factor and non-response errors
...
It is also reffered to as the sampling error, it
is an error that is not within the control of the researcher and it happens because there is
no sample that is a true replica of the population
...
This errors can be minimized by increasing the number of people within
the sample, the greater the sample, the lesser the sampling error
...
This can occur, when the
researcher does not use appropriate sampling technqiues when selecting samples
...
Or situations, where the researcher does not specify the
criteria in which he/she will select individual respondents from the population to make up
the sample
...
This type of error comes about when an element of the sample does not
respond to a measurement instrument for whatever reason
...
2
...
The Sampling Process
This involves a number of activities:
1
...
The definition in any particular case will depend on the study objectives
...
g
...
A welldefined universe specifies
What constitutes the units of analysis;
What the sampling units will be;
What is the specified designation of which units are to be covered;
The boundary, that is, what time period the data refers to
...
Establishing the sampling frame/working population
A sampling frame is a list of all elements in a population from which a sample may be drawn
...
This can be geographic, or listing of
all members of a particular population e
...
electoral register, telephone directory, club
membership, student enrolment list, or a Business Association membership roster
...
3
...
A decision
is made on whether to use probability or non-probability sampling procedures
...
Determining the appropriate sample size
This is influenced by such factors as the level of data precision, degree and nature of analysis,
funds available, time, extend of homogeneity of population elements
...
Execute the sampling process
Apply the operational procedures to select the sample
...
2
...
Sampling Designs:
A sample design is a definate plan for obtaining a sample from the given populations
...
A
sample design is determined before the data is collected and researchers must prepare a
sample design which should be reliable and appropriate for their study
...
Probability sampling technqiue this is where the population have a known probability of being
included in the sample because the number of people in the population is known, this kind of
sampling technique includes a sampling frame
...
Hence, respondents can be
selected through random selection
...
With these kinds of population cases are non-randomly selected
...
2
...
1
...
Includes the following:
Simple Random Sampling
Each sample element in the sample frame has a known and equal probability of selection
...
Simple random sampling is
the purest form of sampling and does not require any prior knowledge or assumption about the
population profile
...
Unique identification numbers are assigned to each study object in the frame sequentially
...
Balls, chips etc can be used
...
Systematic Random Sampling
Unlike in simple random sampling where a random number is drawn each time to select a
sample element, systematic sampling needs to pick only one random number throughout the
process
...
Study objects are then systematically
selected after a fixed interval for sample inclusion after an initial random start
...
Stratified Sampling
The population is divided into subgroups or strata, each to be as homogenous as possible in
terms of the desired characteristic e
...
age or income
...
These samples are then aggregated to form a
stratified random sample
...
Stratified sampling can take two forms:
a) Proportionate stratified sampling
In proportionate stratified sampling, the number of sample elements selected for each stratum
is in direct proportion to stratum size, relative to population size
...
If divided into three strata, given
that sample size is 10% of the population, each stratum should yield 10%of its own
composition
...
Illustration
In a population of 1000 wholesalers, a sample of 100 wholesalers is selected
...
These form the strata and the sizes
are 70, 400 and 530
...
b) Dis-proportionate stratified sampling
The number of sample elements selected from each stratum is not proportionate to its stratum
size
...
Dis-proportionate stratified samples may over or under represent certain strata in the sampling
frame
...
Sometimes the strata in the
sampling frame must be disproportionate in order for the final sample to be proportionate
...
However, the sampling method is more complex, requiring different sample sizes for each
stratum
...
2
...
2
...
There is no exact
way of determining the representativeness of the chosen sample
...
It is best for selecting sampling elements for exploratory research but
not descriptive or causal research
...
It is
commonly used in test marketing, and also to select stores for trying out a new display
...
Quota Sampling
Sample elements are selected on the basis of one or more known parameters (control
characteristic of the population
...
g
...
Usually, each fieldworker is assigned quotas of the number of units (usually persons) to include
according to the characteristic, for which they are to qualify an individual before interviewing
him/her
...
The interviewer should find enough people to
meet the quota
...
58
Snow Ball Sampling
This is also referred to as chain-referral sampling
...
Usually, these are groups that are normally hard to locate
...
For instant, if you are trying to locate a group of people who have a
certain stigmatized social condition such as being homeless, having a certain illness, or sharing a
common faith, this method can be most appropriate method to recruit new participants
...
2
...
Sample Size determination
The major considerations include:
i
...
With large populations there is
less precision
...
Nature of the research design- is it qualitative or quantitative, if qualitative, the
exploratory research design may be used hence, the need for smaller samples,
whereas with survey or descriptive research design, large populations are required
to enable generalization of data
...
Statistical analysis to be used- large samples require sophisticated analysis
iv
...
Number of traits to be measured- the more the traits or variables to be observed the
greater the need for a larger sample size
vi
...
Population size- if small and manageable, then do a census instead of sampling
viii
...
59
5
...
Summary
Sampling is very important step in the research process, it can make or break
the research process
...
Sampling techniques are of two
categories: probability and non-probability techniques
...
5
...
1
...
Researchers must make every effort to minimize sampling erro and sampling bias
that occur due to researchers choices
...
3
...
Activities
From the identified research problem, identify the ways in which you can draw representative
samples through the probability sampling technique
60
5
...
3
...
A
...
G
...
John Wiley and sons,
New York
...
T
...
Butterworth-hiennemann Oxford
Blankenship, A
...
Breen, G
...
and Dukta, A
...
Contemporary publishing group Inc
...
Boyd, H
...
Westfall, R
...
F (1990) Marketing Research texts and cases
...
Richard D Irwin Inc
...
Baker, M
...
(1991) Research for marketing, The Macmillan Press Ltd Hong Kong
...
3
...
Self-Test Questions
Compare and contrast the probability and non-probability sampling techniques
Discuss ways in which researchers can randomize the samples so as to have proper
sample/population representation
Discuss the advantages and disadvantages of probability and non-probability sampling
techniques
61
6
...
Data Collection
In order to respond to the research questions, data has to be collected
...
The collected data must be reliable
and valid in responding to the research questions
...
Data is a key part
of the research plan and often represents the biggest cost in primary research
...
6
...
Objectives
By the end of the lecture, you should be able to:
Explain the difference between primary and secondary data
Identify the typical sources of secondary information
Identify appropriate sources of primary data
Describe the various methods of obtaining primary data
6
...
Types of Data
There are two basic types of data: primary and secondary data
...
It is first hand collected data to solve
the underlying research problem
...
62
6
...
1
...
Internal sources refer to the secondary data that is already contained in the companys datases
that can be used to analyse a situation
...
Whereas external sources refer to the secondary data that collected by other individuals or
organizations on behalf of the organization
...
mostly this data is gathered by the
organization to enable inform it on marketing intelligence
...
6
...
2
...
Observation research- this refers to simply observing the consumers without communicating
with them
...
For observation to be a valid way of collecting data, it must ascertain the following conditions:
the desired information should be in such a way that it can be inferred from the observations of
the subject behavior, the behavior must be of interest, frequent, repetitive, or predictable and
the behavior of interest must be of relatively short durations of time
...
63
a) Natural versus Contrived Observations
Natural refers to counting naturally occurring events
...
b) Disguised versus Undisguised (Open) Observations
Disguised refers to the process of monitoring people, objects, or occurrences that do not even
know they are being watched, much less why such as through CCTV cameras in the retail stores
...
c) Structured versus Unstructured Observations
Structured refers to a study in which the observer fills out a questionnaire-like form, or counts
the number of times an activity occurs
...
It is usually used when the researcher doesn't know much about the behavior or
subject of interest; used in this way, it is a preliminary study to be followed later by a structured
study or observation
...
However, past behavior can
be indirectly observed by reviewing some record or evidence of past behaviors or lack thereof
(such as how much of a product was "unused" over some period of time)
...
Thus, they are used when the situation
allows and when the machines are available
...
Human Observations
This refers to people (rather than machines) watching other people
...
One-Way Mirror Observations - The practice of watching unseen from behind a one-way mirror
...
64
Shopper Patterns - refers to drawings that record the footsteps of a shopper through a store
...
Also used to study
the effect of music on shopper behavior
...
Trace analysis - refers to data based on physical traces of past behaviour such as what
newspapers people read, cooking fats used (gathered through garbage research)
...
Content Analysis attempts to
determine what is being communicated to a target audience by objectively and systematically
describing the communication's content
...
Humanistic inquiry is a non-empirical method of recording
and analyzing market phenomena using two diaries (or logs) supplemented by participant
observations using audiotape or videotape recordings, artifacts (e
...
, shopping lists or garbage),
and supplemental documentation (e
...
, magazine articles, health records, survey data, or
census reports)
...
It shows the process by
which the researcher has come to understand the phenomenon being studied
...
Outside auditors then judge whether interpretations are made
in a logical and unprejudiced manner from the data gathered and the rationale employed
...
ii
...
These include Traffic Counters -machines used to
measure vehicular flow over a particular stretch of roadway, scanners in supermarkets,
pedestrian turnstiles, people meters attached to TVs to monitor viewership, or even
videotaping customers in stores
...
Activation - refers a person's feeling of arousal,
inner tension, or alertness
...
The sight of a product or
advertisement can activate RAS [or provoke the arousal processes of the RAS], which increases
the person's processing of information
...
It detects arousal levels that are created by outside stimuli such
as advertisements
...
GSR usually is used to measure
stimulus response to advertisements
...
Pupilometer - measures changes in pupil dilation
...
But, it really measures some unknown combination of arousal,
mental effort, processing load, and anxiety
...
Voice Pitch Analysis - The study of changes in the relative vibration frequency of the
human voice in order to measure emotion
...
Used in package research, to predict brand preference, and to determine
predisposition to buy a product
...
However, the validity of Voice Pitch studies is questionable
...
Arousal produces adrenaline, which enhances the activation process via a faster heart
rate, increased blood flow, an increase in skin temperature and perspiration, pupil dilation, and
an increase in brain wave frequency
...
66
6
...
Data Collection instruments
An instrument is a tool that we use to implement a project or a purpose and is usually used in
science because of its precision in obtaining the results
...
The purpose of a tool or research
instrument in research is to measure the research variables
...
The type of tool to be used depends on the design and type of study
...
Types of Data Collection Instruments
Questionnaires
Interviews
Focus group discussions
Projective techniques
Tests
Documents analysis
6
...
1
...
A questionnaire
consist of a number of questions printed or types in definite order on paper
...
A
questionnaire should be well constructed as it is considered the “heart” of the survey
...
Should be brief with non-ambiguous clear
questions, it should be objective and not ask leading questions, and most importantly is should
be attractive in appearance, neatly arranges, and clearly printed
...
3
...
1
...
Usually for quantitative data analysis
Unstructured or open ended questions: these questions give the respondent complete freedom
to respond
...
They are usually for
qualitative data analysis
...
In
cases where the respondents want to probe for more information from certain respondents in
the population he/she can follow-up on the questions that are used to probe for more
information from the relevant sub-groups only
...
For instance, if you wanted to do a study on customers’ perception on your services, you
can ask a set of questions that share the same response categories such as: strongly agree,
agree, indifferent, disagree and strongly disagree
6
...
1
...
Rules for constructing a questionnaire
-
Before constructing a questionnaire, list the objectives that you would like the
questionnaire to accomplish
...
Terms like severally, frequently, or
most likely have no precise meaning and should be avoided
...
E
...
the term customer
-
Double-barred questions must be avoided; every question should assess one idea
-
Misleading or biased questions should be avoided
...
E
...
have you stopped buying goods and services on credit?
-
Avoid psychologically threatening questions e
...
are you worried about the
consequence of contracting HIV?
6
...
1
...
Steps and consideration in questionnaire construction
1
...
This can be put in a summary form
...
At this
stage, the researcher also decides on the analytical methods to use so that he gathers data in the form
appropriate for the desired analytical technique
...
Select the survey method
The type of questionnaire to use will depend on the method of administration to employ, i
...
, whether
personal interviews, telephone interviews, or mail interviews will be used
...
3
...
Questions/responses are classified into two: open-ended and closedended question
...
Three Types of Question Format
Free response-questions vary on the amount of freedom they give to the respondent
...
Then the
interviewer starts to probe with informal interviewing to pursue the subject further using such
phrases as anything else•, why do you say that•
...
g
...
Closed- ended questions- questions that asks the respondents to choose from a list of answers
...
Multiple Choice (Multi-chotomous) Questions - questions that ask a respondent to choose among a
list of more than two answers
...
Rating Scales- These are also types of scaled response questions such as Likert Scales
...
Effective rating scales use descriptors with
clearly understood measures such as frequency: always, usually, seldom, rarely, never… etc
...
Rankings scales- these are the kinds of questions that force respondents to rank or order a set f
attributes as regards to the concept
...
For instance, the researcher would be interested in identifying the presence of certain
aspects of the scope of the study
...
g
...
4
...
There are several guidelines concerning wording:
use simple language, use familiar vocabulary, keep questions short, be as specific as possible, avoid
double-barreled questions (covers two issues at once), avoid leading and loaded questions (a loaded
question suggests socially desirable answers or is emotionally charged), avoid implicit assumptions,
avoid estimates as well as burdensome questions taxing the respondents memory
...
70
Survey questions can be threatening or non-threatening
...
The main problem with non-threatening questions is memory
...
When asking threatening questions, use a randomized response methodology; deliberately load the
question; use a long time frame and embed the question
...
5
...
Essentially,
the questionnaire should have an introduction, body and conclusion
...
Questions
dealing with the same topic should be placed close together
...
Determine the physical characteristics of the form
There are thus three macro issues in the overall questionnaire design namely:
a
...
Order effects
c
...
Appearance is very critical for self-administered questionnaires where respondents
are not motivated
...
It’s important to
consider all aspects influencing appearance including quality of paper ant printing
...
For open-ended questions,
leave enough space for respondents to fill in answers
...
For example, one can publish it in a booklet rather than stapling many pages together
...
g
...
In summary, the
questionnaire aesthetics is maximized through the following considerations: use booklet form,
appearance is important, do not crowd the questions, use large and clear font type, number the
questions, do not split a question between two pages, use parallel columns when asking identical
questions about multiple brands
...
Pretest, revise and final draft
Normally faults emerge in the field and therefore to eliminate this, pre-test the questionnaire in a small
number or typical respondents i
...
Before distributing it on a major scale, test it on a group that is
comparable in make-up to the intended audience
...
Therefore there are two pretests:
i
...
The process
The pre-testing indicates:
If the questionnaire is understandable
Whether the interviewers can administer it well;
Nature of information gathered-free response answers;
Time required to administer the questionnaire;
Any problems encountered in the administration
...
After pre-testing, revise the questionnaire and write a final draft
...
3
...
Interviews
An interview is an oral questionnaire
...
To use this method, the
interviewer must be skilled to gather valid information
...
The main requirement for successful data collection using this technique is the ability to approach
identified respondents to persuade them to participate in the study
...
Interviews are of two main kinds: personal interview or mail interview
6
...
2
...
Personal interviews
This is a survey that gathers information through face-to-face co tact with individuals
...
72
Mall-Intercept - Interview consumers in shopping malls or other high-traffic locations (usually done in
public areas or interviewees may be taken to some nearby private area)
...
This method is relatively simple, yet effective & efficient
...
The
method is however very expensive and time consuming
...
3
...
2
...
It is usually shorter
than personal interview
...
Central Location Telephone Interview - Interviewers make calls from a company facility to reach and
interview respondents
...
Computer Assisted Telephone Interview (CATI) - Central location telephone interviewing in which the
interviewer reads questions from a computer screen and enters respondents answers directly into the
computer
...
This method allows the interviewer to input
responses directly into the data set which avoids the cost and processing errors associated with manual
coding and entering of data into the computer sometime after the interviews generated the raw input
data on paper
...
The recorded voice of an interviewer asks questions and respondents answer by pressing
numbers in their push button phones
...
The computer asks questions, leaving blank tape in between to record the answers
...
6
...
3
...
In FGDs the researcher should have specific topics to be discussed and a recording
tape of the discussion made
...
The technique is a qualitative technique involving
a small group of people talking spontaneously and freely on themes considered important to an
investigation or research, it has a moderator and a researcher and usually it is for ideas that the
73
researchers need varied responses about
...
Focus groups should however, not be used in cases where quantitative data is needed, and should not
be used as a decision-making tie breaker or for lobbying purposes
6
...
3
...
Types of Focus Groups
Exploratory Groups (problem development) - Focus groups that aid in the precise definition of the
problem; are used in pilot testing of questionnaires or new products; are used to generate hypotheses
for testing; or are used to develop concepts for future research
...
Experiencing Focus Groups (observing consumers in action and hearing motivations for purchase) Focus groups that demonstrate or describe in detail the use of the product
...
3
...
2
...
Preparation:
Preparation involves providing the setting, usually in a conference room and with an available taperecorder to record the information; recruiting the participants; and setting the focus group timing
usually 90 minutes, from which 10 minutes are devoted to introductions and explanations of procedures
while 80 minutes of the time are for the session with the moderator taking no more than 25% of the
time
...
Selecting the moderator:
This refers to the person tasked with the duty to lead the focus group
...
The person should be
provided with a discussion guide which outlines the topics to be covered during the focus group
discussion and which sets the framework for the moderator
...
The most common form of report by the moderator is the Cut and Paste Technique- where the
researcher has the sessions transcribed, reviews them looking for common threads or trends in
response patterns, and cuts and sorts similar or matched patterns of responses into homogeneous
groups
...
Next is a two or three page summary of findings and
recommendations on each topic (which is the main body of findings)
...
Subsequent topics are then covered in similar fashion
...
3
...
Projective Techniques
Ways of tapping respondents' deepest feelings by having them "project" their own feelings onto
unstructured situations
...
Types include:
1
...
This projective technique is
practical and effective, and often used to select brand names, advertising campaign themes, and
slogans
...
Sentence and Story Completion - Tests in which the respondents complete sentences or stories (i
...
,
structured scenarios) in their own words
...
They can be used in conjunction with word
association
tests,
and
they
are
most
reliable
...
Cartoon Tests - Tests in which the respondent fills in the dialogue of one character in a cartoon
...
Sometimes they also are used to ascertain what function is being performed by a given
attitude
...
Third Person Techniques - Ways of learning respondents' feelings by asking them to answer for a third
party; such as completing statements that begin with "your neighbor
...
" The
respondent is expected to transfer his /her attitudes to the third person
...
5
...
75
6
...
5
...
A test is an ainstrument designed to
describe and measure a sample of certain aspects of human behavior
...
They can be classified as follows:
Performance tests- administered individually and require that the subjects manipulate objects or
mechanical apparatus while their actions are being observed and recorded by the examiner
Paper and pencil tests- administered in groups and require the subjects to mark their responses on a
prepared sheet
Power tests- they have no time limit and the subject attempt progressively more difficult tasks until they
are unable to continue successfully
Timed or speed tests- basically involve the lement of power but in addition they limit the time the
subjects have to compete the tasks
Standardized tests- the content, administration, and scoring have been set in one patter so that those
taking it will do so under similar conditions
6
...
6
...
The analysis of content is a central activity whenever one is concerned with
the study of the nature of the verbal materials
...
It can
also be subtle when the researcher makes a study on the attitude such as a feature analysis on
consumer behavior or trends
...
4
...
Secondary data is
data that had been previously gathered for some other purpose, other than the current research
problem under investigation
...
Primary data is data generated from a survey, observation, or experiment just to solve the particular
problem under investigation
...
Each of the techniques has various
strengths and limitations, and proper care must be made when deciding on the primary data collection
method to be used
...
There
are several important considerations to be made when constructing the research questionnaire, such as
wording, length, mode of questionnaire administration, as well as its physical characteristics
...
4
...
Note
The type of research design determines the type of data to be collected and the means or
techniques through which that data is collected
...
6
...
2
...
You are interested in determining the consumer perception of customer care services in
commercial banks in Kenya
...
77
6
...
3
...
A
...
G
...
John Wiley and sons, New
York
...
T
...
Butterworth-hiennemann Oxford
Blankenship, A
...
Breen, G
...
and Dukta, A
...
Contemporary publishing group Inc
...
Boyd, H
...
Westfall, R
...
F (1990) Marketing Research texts and cases
...
Richard D
Irwin Inc
...
Baker, M
...
(1991) Research for marketing, The Macmillan Press Ltd Hong Kong
...
4
...
Self-Test Questions
Compare and contrast the primary and secondary data?
Providing relevant examples discuss the conditions in which it is most appropriate to collect primary or
secondary data
How does the choice of a research design influence the choice of data collection technique?
78
7
...
Most of the data collected through questionnaires if not checked
against the threshold of reliability, validity, and usefulness, it may yield findings that are not
generalizable, and that do not reflect the reality on the ground
...
It is important that the developed questionnaire yield data that is useful in deriving
answers to the research questions hence providing authentic information that guides decision makers
...
Measurement is the
process of assigning numbers to objects in accordance with specific rules to represent quantities of
attributes
...
This lesson
is therefore dedicated to discussing these concepts
...
The criteria of choosing a scaling technique are also highlighted in the
lesson
...
1
...
2
...
Thus, measurement is a procedure used to assign
numbers that reflect the amount of an attribute possessed by an event, person, or object
...
First, numbers permit
statistical analysis of the resulting data
...
The most important aspect of marketing is the specification of rules for assigning
numbers to characteristics
...
e
...
7
...
1
...
Identify the Construct (Concept) of Interest:
A construct is the invented name of a property or concept
...
Such as: Marital Role, Social Class, etc
...
2
...
This is similar to a
dictionary definition that simply states the central idea (or concept) under study, which distinguishes it
from all other similar concepts
...
This definition includes the process for
assigning a value to the concept, where the process is stated as a set of measurement rules to be used in
the investigation
...
Therefore, the operational definition of the concept gives
meaning to that concept by spelling out what the researcher must do to measure it
...
Develop a Measurement Scale:
A scale is a set of symbols or numbers so constructed that the symbols or numbers can be assigned by a
rule to the individuals (or to their behaviors or attitudes) to whom the scale is applied
...
80
7
...
Levels of measurement of scale
There are four levels of measurement: nominal, ordinal, ranked, and interval
...
There are certain statistical analysis which are only meaningful
for data which are measured at certain measurement scales, for instance variables such as gender have
dichotomous responses who are either male or female, for such data, the only possible type of analysis
is mode and median because the data is categorical and not continuous in nature
...
They produce non-metric data, such as male (1) and female (2)
that simply identifies objects, events, or groups
...
In marketing research, nominal scales are used for identifying respondents,
brands, attributes, stores and other objects
...
The mode is the measure of central tendency
...
The only measure of average that can
be used is the mode because this is simply a set of frequency counts
...
The most likely would be the Chi-square test
...
It can tell nothing about the form of that relationship, where it exists,
i
...
it is not capable of establishing cause and effect
...
They produce non-metric data that provide information about the
relative ranking of some characteristic possessed by an event, object, or group, such as quality ranking,
social economic class, and so on
...
However, the amount of difference between responses is unknown
...
In marketing research, ordinal scales are
used to measure relative attitudes, opinions, perceptions, and preferences
...
For example, Company X has
three main products: A, B and C
...
The Mode and the Median are both used as measures of central tendency
...
( 1 being most preferred, and
5 least preferred)
Omo
Ariel
Sunlight
Toss
Gentle
From such an example, the researcher knows the order of preference but nothing about how much
more one brand is preferred to another, that is there is no information about the interval between any
two brands
...
In
addition, positional statistics such as the median, quartile and percentile can be determined
...
The two main methods are Spearman's Ranked
Correlation Coefficient and Kendall's Coefficient of Concordance
...
Consider again the ranking of washing detergents example above
...
The resultant coefficient
takes a value in the range 0 to 1
...
It is more likely that an answer somewhere between
these two extremes would be found
...
In this
way, they show order and differences between responses
...
Both the zero point and the unit of measurement are therefore arbitrary
...
The interval or cardinal scale has equal units of measurement, thus making it
possible to interpret not only the order of scale scores but also the distance between them
...
This of course has
implications for the type of data manipulation and analysis we can carry out on data collected in this
form
...
An interval scale contains all the information of
an ordinal scale, but it also allows to compare the differences between objects
...
The difference
between "1" and "2" is equal to the difference between "3" and "4"
...
And, three is a constant or equal interval between
scale values
...
[This limitation is mathematically due to the arbitrary, man-made zero we assigned to
the scale
...
We cannot speak of
50°F being twice as hot as 25°F since the corresponding temperatures on the centigrade scale, 10°C
and -3
...
Interval scale uses question formats such as rating and ranking scales such as Likert scales, or scales that
are aimed at assessing the extent of agreeableness, frequency, extent of likelihood etc
...
This means that the "magnitudes" between any points on
the scale can be compared arithmetically
...
In ratio scales, we can
identify or classify objects, rank the objects, and compare intervals or differences
...
Not only is the difference between 2 and 5 the same as difference
between 14 and 17, but also 14 is seven times as large as 2 in an absolute sense, someone weighting 70
kg is said to be twice as heavy as one weighting 35kg
...
All statistical techniques can be applied to
ratio data
...
7
...
1
...
A
...
For example, the
researcher can ask the respondents whether they prefer Fanta Orange or Citrus
...
The data
obtained are ordinal in nature
...
The respondents can prefer Coke to Sprite or Fanta to Krest
...
Under the assumption of
transitivity, it is possible to convert paired comparison data to rank order
...
Paired comparison scaling is useful when the number of brands is limited, since it requires direct
comparison and overt choice
...
Besides, Paired comparisonscaling bear little resemblance to the market situation, which involves selection from multiple
alternatives
...
83
Rank order scaling is also a comparative scaling technique in which respondents are presented with
several objects simultaneously and asked to order or rank them according to some criterion
...
Rankings can then be
obtained by asking the respondents to assign a rank of number "1" to the most preferred brand, number
"2" to the second most preferred and so on, until a rank of "n" is assigned to the least preferred brand
...
The
major disadvantage of the technique is only ordinal data production
...
For example, the respondents are given
20 attitude statements on product satisfaction, ranging from "most highly satisfied to least highly
satisfied"
...
Magnitude estimation: In magnitude estimation technique, numbers are assigned to objects such that
ratios between the assigned numbers reflect ratios on the specified criterion
...
Then they assign number between 0 to 100 to each statement to
indicate the intensity of their agree or disagree
...
The respondents are asked to "think out loud" and verbalize anything going through their
heads while making a decision or performing a task
...
Protocol
analysis has also been employed to measure consumer response to advertising
...
This technique respects stimulation from the researchers to responds
...
At these times, the researchers will say "remember
to say aloud everything you are thinking"
...
This
record of the respondents verbalized thought processes is referred to as a protocol
...
Support argument- Support the claim made
Counter argument- Refute the claim made
Source derogation -Negative opinion about the source of the message company
Typically, protocols are incomplete
...
The researchers must take the incomplete record and infer from it a measure of the
underlying cognitive response
...
Non- comparative scaling techniques
Non-comparative scaling techniques are one of two types of scaling techniques in which each stimulus
object is scaled independently of the other objects in the stimulus set
...
They do not compare
the object being rated either with another object or to some specific standard
...
Continuous rating scale refers to as a graphic rating scale, the respondents rate the objects by placing a
mark at the appropriate position on a line that runs from one extreme of the criterion variable to the
other
...
The form of the Non-comparative scaling techniques may very considerably
...
The following are three versions of a Non-comparative scaling
techniques
...
Strongly
Version
2:
Strongly
disagree
...
Strongly
agree
1
2
3
4
5
6
7
8
9
10
The respondent has provided the ratings, the researcher decides the line into as many categories as
desired and assigns scores based on the categories into which the rating fall
...
Although continuous scale is easy to construct, it provides little new
information
...
-
Version
3:
Itemized
rating
scales
These are measurement scales having numbers or brief descriptions associated with each category
...
Itemized rating scales are widely used in
marketing research and form the basic components of more complex scales, such as multi-item scales
...
Typically, each scale
item has five response categories, ranging form "strongly disagree" to "strongly agree"
...
This means that the items making up a
Likert scale are summed to produce a total score
...
Typically, each scale item will have 5 categories, with scale values ranging from - 2 to +2 with 0 as
neutral response
...
In a typical application, respondents rate objects on a
number of itemized
...
It has been
widely used in comparing brands, products and company images
...
Staple scales-This is a unipolar rating scale with ten categories numbered from –5 to +5,without a
neutral point(zero)
...
Respondents are asked to indicate by selecting an
appropriate numerical response category how accurately or inaccurately each term describes the object
...
For example, respondents
may be asked to evaluate how accurately each word or phrase describes a certain supermarket in terms
of service quality
...
86
C
...
The characteristic
to be measure is frequently called a construct
...
A theory is necessary not only for constructing the scale but also for
interpreting the resulting scores
...
From this
pool, a reduced set of potential scale items is generated by the judgement of the researcher and other
knowledgeable individuals
...
The data are analyzed using technique such as correlations, factor
analysis, discriminant analysis and other analysis
...
The purified scale is evaluated for reliability and
validity by collecting more data from a different sample
...
An example of a scale developed for use in industrial marketing research as below: The following multiitem scale measures the technical sophistication of a product line
...
7
...
Measures of Sound Measurement
A measurement tool in research is any instrument that is used to collect data such as a questionnaire
...
A sound measurement tool is valid,
reliable, and practical
...
Measurement accuracy is fundamental to scale evaluation
...
7
...
1
...
Validity suggests that the
measurement device is substantially free from error, including both systematic (measurement and
sampling biases) and random error
...
The systematic errors result in a constant bias in measurement, it
results from the faults in the measurement instrument or in the measurement process
...
In ensuring validity of the items within a measuring instrument, it is important for the researcher to
operationalize the study variables
...
For instance, the
concept on service quality has been operationalized into five specific indicators: responsiveness,
assurance, tangibility, empathy, and responsiveness
...
The table below shows specific ways
in which a construct can be broken down into:
The Rater Model- Service Quality Dimensions
Dimension
Definition
Reliability
delivering
promises
( Your ability to perform
on the promised service
dependably and
accurately
Assurance ( inspiring The knowledge and
trust
and courtesy of staff, their
confidence)
ability to inspire trust
and confidence
Specific criteria that are Questions
in
the
used to measure the questionnaire
indicator
Timeliness
Does the company
Consistency/regularity
deliver items on time
Accuracy
Staff competence
Respect for stakeholders
Credibility
Confidentiality
Safety and security
Does your organization
instill
trust
and
confidence in their
transactions
Tangibility(
The physical
representing
the representation or
service physically)
images of service
Physical facilities
Equipment’s
Technology
Employees
Communication materials
Are
the
physical
facilities
in
your
organization
appealing?
Empathy ( treating The caring
customers
as individualized attention
individuals)
you provide your
stakeholders
Access to staff, services,
communication
Communication ( clear,
appropriate and timely)
Services appropriate to
stakeholder needs
Individualized attention
Willingness to help
Prompt attention to request
and questions
Problem resolution
Complaint handling
Flexibility
Does your head of
department
provide
individualized
attention
to
your
personal needs
Responsiveness
( Your willingness to help
being willing to help) customers and to
provide prompt service
88
Does your company
provide
prompt
responses to your
request
7
...
2
...
This is a
judgement call by the researcher, which was actually made before measurement as the questions were
designed
...
Thus, there are three "direct" assessments of the validity of a measurement
instrument i
...
Content Validity, Criterion-Related Validity, and Construct Validity
...
It provides adequate coverage of topic
under study
...
It is the extent to which an instrument has been accurate in
measuring what it ought to measure
...
Rather than
the logical methods of identifying experts to estimate the validity, this kind of validity, evaluates the
results of the pilot study against the actual results of the study, to find out the degree of association
between the findings
...
Predictive or criterion-related Validity
This refers to the use of a measure in assessing subjects behavior in specific situations for predictive
purposes
...
For instance, if an instrument measures customer
satisfaction, a customer who is likely to be satisfied, or who scores high in customer satisfaction, most
likely scores high in customer loyalty
...
Internal Validity refers to the question of whether the experimental treatment is the sole contributor
of the observed changes in the dependent variable
...
It refers to the extent to which changes in the dependent variable were produced solely
by the independent variables and not extraneous ( other) variables
...
External Validity - The extent to which causal relationships measured in an experiment can be
generalized to outside (other) persons, settings, and times
...
External validity specifies the time, setting, and persons that the results can be generalized to
...
4
...
1
...
History - variables or events, other than those manipulated by the researcher, that occur between the
beginning and the end of the experiment can affect the value of the dependent variable measured at
the end of the experiment
...
For example, competitors may
deliberately alter their marketing strategies during a test-marketing period to cause confusion in the
market situation
...
The use of control group
reduces history risk
...
Maturation Biological and physical changes in the subjects that are a function of time (such as aging,
hunger, or fatigue) may produce different results over time
...
3
...
Variability among observers
exists; and even for the same observer, his standard of measurement may differ from one stage to the
next
...
Also includes change in question wording from pre to post-test, and change in
method of survey administration (personal vs
...
Usually, control group reduces risk, if cannot
eliminate risk through consistent instrumentation
...
Selection Bias - Important systematic differences may exist between the experimental group and the
population of interest
...
Whenever the test units are allowed to select their own groups, self-selection
bias is created
...
Nonequivalent control groups quasi-experimental
design also reduces
...
to obtain highest possible response rate also reduces threat
...
Mortality - Mortality simply means loss of test units
...
As a result, test units at the beginning stage of the experiment
are markedly different from those at the end of the experiment
...
6
...
This can occur because people often respond differently when they know that their behavior is being
monitored (called the Hawthorne Effect), or they have become more familiar with the testing
procedure
...
A Testing Effect can also result from respondents learning how to respond in
repeated testing in an "appropriate" manner, rather than in a "truthful" manner
...
7
...
Generally, people with extreme characteristics tend to move towards an
average position over time or within a group
...
Thus, regression may be viewed as a special form of maturation,
where the change in the test units is the result of their becoming more normal
...
7
...
2
...
Factors influencing external validity
Laboratory experiments often lack external validity due to:
1
...
2
...
This happens when the separate groups that receive different treatments are systematically different
from each other, just because we allowed some bias in our selection of those groups
...
Rather, the results may be due to some combination (or interaction) of the biases present in our
selection of the various groups and the treatments we administer to the independent variables
...
Reactive (or Interactive) Effects of "Testing" - refers to the fact that respondents tend to change their
reactions due to the learning process they naturally proceed through because of being measured before
exposure to the treatment (called the Pre exposure Measurement or Pretest), and then measured again
on the same questions after the treatment (the Post exposure Measurement or Posttest)
...
And, they are correct; they are tested
on the same points
...
Respondents then often try to answer the questions "correctly" rather than the way
they really feel or think
...
4
...
Reliability
Reliability identifies the stability or consistency of the research results
...
Reliability
in research is influenced by random error
...
Methods of assessing reliability
include: test-retest, equivalent forms, and split-half techniques
...
In this methods,
a test is given to a set of respondents and a re-test of the same test given to the same respondents after
a period of time
...
If the researcher obtains the same results on the two administrations’
of the instrument, then the reliability coefficient will be 1
...
Equivalent form/ Alternative form method
Similar to the test-retest methods, the method also requires two tests with the same people
...
Each of the two tests must be designed to measure the same thing
and should not differ in any systematic way
...
This methods is perceived to be superior
than the test-retest because a respondents memory of test items is not likely to play a role in the data
received
...
In the split halve method, the total number of items is divided into
halves and a correlation taken between the two halves
...
Cronbach-Alpha - computes the mean reliability coefficient estimate for all possible ways of splitting in
half a set of individual responses (within the sample) to each item (measure) in a measurement scale (a
group of measurements of a single construct)
...
Note that here, a Measurement Scale refers to a "set" of individual Scaled Measures that, as a group,
are intended to measure a single concept (or construct) such as lifestyle or social class
...
92
You want a alpha coefficient > 0
...
7
...
Summary
Measurement is the process of assigning numbers to objects in accordance with
specific rules to represent quantities of attributes
...
Scaling involves creating a continuum upon which measured
objects are located
...
The scaling techniques can be classified into comparative and non-comparative scales
...
Non-comparative scaling techniques are
one of two types of scaling techniques in which each stimulus object is scaled independently of the
other objects in the stimulus set
...
Research instruments should also exhibit some characteristics of sound measurement: validity,
reliability and practicability
...
There are two main types of validity: internal validity and external validity and
there are certain ways in which validity is ascertained: face validity, content validity, contrast validity,
criterion related validity, predictive validity among other
...
There are four main ways of
ascertaining reliability: test-test, alternative form, split-half and Cronbach alpha
...
5
...
Note
Relationship between reliability and validity can be understood in terms of the true scores model
...
Thus, prefect validity implies prefect
reliability
...
Furthermore, systematic error may also be present, that is Xs 0
...
Reliability is a necessary, but
not sufficient, condition for validity
...
The set of all conditions of measurement over which the investigator wishes
to generalize is the universe of generalization
...
A test-retest correlation is concerned with whether scores obtained from the
measurement scale are generalizable to the universe scores all time of possible measurement
...
To generalize to other universes, generalizability theory procedures must be employed
...
5
...
Activities
A researcher wishes to compare two hotels, Utalii and Stanley, in the following
attributes:
Convenience of location
Friendly personnel
Value for money
a) Design a Likert scale to accomplish this task
...
c) Design a graphic rating scale to accomplish this task
...
5
...
Further reading
Dillon, W
...
, Madden, T
...
H
...
Peterson, Robert A
...
Sommer, Barbara and Robert, Sommer, (1991) A Practical Guide to Behavioural Research: Tools and
Techniques, 3rd Ed, Oxford University Press, New York
...
S
...
(1990) Marketing Research: Measurement and Methods 5th Ed
...
New York
...
H
...
"Theory and Methods of Social Measurement", in Research Methods in the
Behavioral Sciences, Eds
...
and Ratz, D
...
Quee, W
...
(1999) Marketing research 3rd Ed
...
B
...
E
...
(1998) State of the art Marketing Research 2nd ed
...
USA
...
W
...
and Stasch S
...
7th Ed
...
USA
...
J
...
Thurstone, L
...
, (1927), "A Law of Comparative Judgment", Psychological Review 34
...
5
...
Self-Test Questions
Compare and contrast reliability and validity?
Discuss the following statement, “an instrument can be reliable without it being valid”?
Compare and contrast the comparative and non-comparative scaling techniques?
What should researchers take into consideration when choosing a scaling technique to use?
Prepare a questionnaire using the four levels of measurement
95
8
...
After data
has been collected it is classified as either quantitative or qualitative data
...
This kind of data can only be analysed by
organizing the words or images through content analysis
...
Market research data analysis is therefore a blend
of statistics, psychology, information technology, literature and art
...
1
...
2
...
Thus, data analysis may be defined as the whole process, which
starts immediately after data collection and ends at the point of interpretation of the processing
results
...
8
...
1
...
The following are the steps involved in
data analysis
1
...
It is actually the beginning of detection, correction, and
avoidance of errors occurring as a result of mix-ups
...
Validation and editing
Validation is the process of ascertaining that interviews were actually conducted as specified
...
Editing is the process of ascertaining that questionnaires were filled out properly and completely
...
Missing data-This could result from the loss of a source document, failure by the interviewer to ask
some questions, or failure by a respondent to answer the questionnaire at all
...
Some of the strategies of dealing with
missing data include:
Ignoring the missing data in the analysis-if the data existence is unrelated to any other variable or study
object characteristic
...
This is only
recommended when a case has more than 15%missing data
...
Computing a value wherever missing data occurs- can give a random number (up to 10% of the cases
with missing data on a variable), imputing the typical value
(mean, mode, median) of all the cases, or identify a similar case to the one possessing missing data and
impute the same value found in the similar case
...
Consistency-Ensure that skip patterns were correctly followed
...
97
iii
...
Evidence of cheating could be detected e
...
common pattern of responses
from questionnaires of one interviewer
...
3
...
It allows
transfer of data from the questionnaire to the computer
...
Closed-ended questions are usually pre-coded on the questionnaire (Quantitative data)
...
The coding process of open-ended questions involves the following;
a
...
Consolidate (group) responses into homogenous groups;
c
...
Enter codes for each respondent into a computer for statistical analysis
...
For example, consider the following question:
6a
...
It had too much foam
2
...
It left white marks on the clothes
4
...
5
...
Had an tangy feeling
...
Price is too high for the product
...
Use more appealing colours
...
Cut the price
...
Did not like the advertising approach
...
Lack of colour-responses 2, 4, 8
...
Amount used -responses 1, 5
...
High price-responses 7, 9
...
Texture problem-responses 3, 6
...
Advertising appeal-response10
...
Data entry
This refers to the process of physically entering numeric values into the computer
...
Sometimes the data are first transferred to a computer-coding sheet,
but this often creates errors in the transferal process that may not be caught and corrected
...
This is where the GIGO (Garbage In Garbage Out) principle must be put in mind at all times to
avoid writing a report based on Garbage Results
...
Intelligent Data Entry Systems (Devices)-refers to the logical checking of information by the
computer as the data is entered into the entry device (computer), which is programmed to check for
and avoid certain types of common errors at the point of data entry
...
Sometimes a second computer is attached to the data entry device
to perform this checking routine
...
Dumb Data Entry-In this case, there is no automatic checking by the computer
...
5
...
This final validation check of the data is absolutely necessary to avoid having to come back
to the original questionnaire or raw data too many times to correct errors when you are in the middle of
the analysis
...
If a checking program is available, a final computerized error check of
tabulated data is then performed using Error Checking Routines
...
The routines check for the
presence of various conditions that may have been violated
...
Data Processing
This involves the proper selection of the analytic procedure and subsets of data that are to be used in
particular analyses; selection of the final versions of the variables to be used; decisions about what
statistics are to be calculated; setting up the runs to produce these statistics and eventually submitting
the setting to the processor for computation/processing
...
Interpretation of results
This involves deriving some understanding from the output relative to the subject matter of the
research and based on the derived understanding make conclusions
...
3
...
Such findings are not always computable by arithmetic relations
...
It is a flexible qualitative technique of analyzing text data
...
It refers to the subjective interpretation
of the contents of text data through systematic classifications process of coding, and identifying themes
and patterns
...
Assertion analysis- provides the frequency with which certain objectives are characterized in a particular
way
...
4
...
In choosing the statistical approach to use in data analysis, a
number of important considerations need to be made:
i
...
ii
...
Scale of measurement e
...
nominal scale use mode while interval scale uses mean
...
Parametric versus non-parametric statistics-interval or ratio scaled data with large sample size can fit
a normal distribution hence parametric statistics are ideal
...
Characteristics and expertise of the researcher
...
4
...
Types of analysis
In computing and interpreting quantitative analysis, there are two main techniques of summarizing and
analyzing the data: descriptive analysis and inferential analysis
8
...
1
...
Descriptive Analysis
Descriptive analysis refers to the process of transforming raw data into a form that will make it easy to
understand and interpret: rearranging, ordering and manipulation of data to generate descriptive
information
...
Descriptive statistics include sample size, mean, minimum, maximum, standard
deviation, variance, range, and sum, standard error of the mean, and kurtosis and skewness with their
standard errors
...
Measures of Central Tendency describe the centrality of a frequency distribution and include:
Mean - the sum of the values for all observations of a variable, divided by the number of observations
...
The
mean only has "real" value when calculated from interval or ratio scaled (metric) data
...
Arithmetic mean
Median - The observation below which 50% of all observations fall
...
Mode - the value that occurs most frequently
...
The mode can be calculated for and represent any type of data
...
It indicates how spread
out the data is
...
Sample Standard Deviation for a variable (S) - the square root of the sum of the squared deviations of
all observations from the mean of those observations, divided by the degrees of freedom (which is the
number of observations minus 1)
...
In other words, it is the average of the squared deviations of observation values from the mean of the
distribution
...
101
8
...
1
...
Inferential Analysis
The essence of conducting quantitative analysis of sample populations is so as to make inferences or
extrapolations about the entire population using the small sample size available to the researcher
...
That is, we want to be able to confidently
assume that the population would give us virtually the same responses as did the sample
...
When these mathematical differences between the groups' responses concerning a variable are
"statistically" different, the presence of statistical difference suggests that we can not generalize the
sample results to the population of interest, or to any another group of interest
...
However, if a difference "is" statistically significant, a marketing manager can legitimately choose to
ignore the difference, if the absolute mathematical difference is so small that the manager's financial
risk of relying on the numbers is almost nil
...
It is only large and significant differences that must
never be ignored by marketers
...
4
...
3
...
We also do inferential analysis to identify if there are differences, or statistically
significant differences between two groups
...
Univariate statistical analysis is used when one wants to generalize from a sample about one variable at
a time e
...
T-Test for independent samples or dependent samples
Bivariate statistical analysis involves statistically describing the relationship between two variables at
one time e
...
correlations, regression analysis
Multivariate statistical analysis involves simultaneous investigation of more than two variables
...
g
...
102
8
...
Hypothesis testing
Hypotheses are specific assumptions or theories that a researcher or manager makes about some
characteristic of the population under study
...
" It is a
statement about the expected relationship between observable or measurable events or expected
differences between groups under investigation
...
Analysis that are used to establish relationships between variables include:
correlations and regression analysis
...
Parametric tests are analysis done when the
population under investigation assumes some characteristics which include: the population has a
normal curve, the number of respondents is large preferably more than 30 and when the analysis can be
evaluated using various diagnostic tests
...
Hence, if the researcher wants to
determine relationships between variables within a population that makes assumptions of normality
and large sample size, the appropriate correlations to be conducted include Pearson correlations
...
Correlations are done to establish associations whereas regressions are conducted to establish
predictions
...
A positive correlation shows the direction of
the association, meaning an increase in one unit of a measure of variable will result to an increase in
another unit of the other variable
...
The close the association is to one, the greater
the relationship and vice versa
...
Hypotheses also aim at identifying if there are differences between two or more groups
...
T-Tests are conducted if you are establishing
differences between two groups and ANOVA tests are conducted if you are establishing differences
between more groups than two
...
To illustrate this, T-Test for independent samples can be conducted in marketing research in
instances where a marketer wants to establish if there are significant differences between two separate
groups such as males and females
...
g
...
For
such a study, a T-Test for independent samples could be conducted to establish if there are differences
between the groups
...
In such a case, since the same participants’ are used,
the appropriate statistic to conduct would be T-Test for dependent samples
...
Market researchers conduct these analyses to identify if there
are significant differences between the groups so that the analysis can enable them in making marketing
decisions on how much capital to invest, what kinds of promotional techniques to utilize for different
groups, what pricing techniques to use for different groups etc
...
5
...
Types of Hypothesis
There are "basically" two types of hypotheses:
Null Hypothesis and Alternate Hypothesis, these are further classified as directional and non-directional
...
It is what could be
expected knowing no other information; it is also referred to as the “Default state of events”
...
Hence knowing no other information, you would not expect any two variables to be related or any two
groups to be different; until a study is done that either approves or disapproves that assumption
...
An example of a null hypothesis testing for differences:
H0: There is no significant difference between males and females consumer buying behavior in the retail
industry
104
Alternate Hypothesis
This is also referred to as the “Researchers hypothesis”
...
Based on what is already known from literature, observations, or intuition, a
researcher can aim at establishing if there are actual relationships between two variables or differences
between two or more groups
...
An example of an alternate hypothesis for relationship
H1: There is a significant relationship between advertising costs and sales revenue in clothing
merchandise
...
An example of an alternate hypothesis for differences
H1: There is a significant difference between males and females consumer buying behavior in the retail
industry
Other than, being null and alternate, hypothesis can also be directional versus non-directional
...
For instance in the examples provided above, the researcher could
make the following assumption:
H1: There is a significant positive relationship between advertising costs and sales revenue in clothing
merchandise
...
Meaning, in this case, he/she assumes that increase in
advertising costs will increase sales revenue
Or a directional alternate hypothesis to test for differences could be as follows:
H1: Females are likely to purchase on impulse more than males in the retail industry
...
Non-directional hypothesis are hypothesis that do not state the direction of the hypothesis
...
Note: the direction of the hypothesis is only stated in the alternate hypothesis
...
105
8
...
2
...
Each hypothesis should show identify one independent and one dependent variable
It should be reasonable and consistent with the existing body of knowledge
It should be statistically or numerically testable and verifiable, meaning, hypothesis testing is only done
for quantitative analysis and not qualitative analysis
...
However, it is important to note that hypothesis testing is not equal to decision making
...
Also, the hypothesis test is limited in that it does not explain reason for the differences
or relationships which could be accounted for by sampling fluctuations or sampling biases
...
Hence, when interpreting analysis obtained through hypothesis testing, caution needs to exercised
...
5
...
Steps for hypothesis testing
1
...
Knowing no other information, there is no
relationship between variables, or no differences between groups
...
The statement is
stated in a manner that shows statistical significance, in social sciences the accepted statistical
significance is a p-value of less than
...
A p-value is the value that denotes the probability of rejecting a null hypothesis when it is true
...
Hence a P-value of 0
...
In more stringent
fields like medical fields, the p-value is placed at 0
...
Hence, it is more prudent to state that a
certain promotional technique has no relationship with sales, rather than stating that there is a
relationship, and then its’ found to be false
...
106
2
...
If relationships, the appropriate test statistic would be a
correlation or regression, if differences, the appropriate test statistics would be T-Test or ANOVA
3
...
Researchers most often choose a significance level (called alpha) of
...
The specific purpose for setting this criteria (alpha level) is to determine whether the difference
between the observed result (such as an actual sample mean score for all responses for a variable) and
its expected value (which is the mean score stated in the null hypothesis) could have occurred by chance
(a sampling error) less than 5 times out of 100
...
This
result would be shown by statistically developing an "actual" significance level (called Pvalue) of less
that the cutoff point (called alpha) that we chose as our criterion for testing the null hypothesis
...
Test of Significance
...
In our "test of significance," if our calculated P-value (actual significance level) is less than
...
e
...
Meaning, the differences within the sample and the population is
actual, and it is it not random
...
This significant
result indicates that the probability that the occurrence of the observed sample result is due to chance
(a sampling error) is less than 5%
...
In other words, we accept our alternate hypothesis because there is too great a probability that the
acceptance of the null would be committing a Type II error (or Beta error), which is defined as the
acceptance of a false null hypothesis (of "no difference" in this example)
...
5
...
1
...
It is the incorrect
rejection of a false null hypothesis
...
More
simply stated, it mean stating that there is a relationship or difference when it actually does not exist
...
g
...
10
...
1, would mean there is a 10% chance of stating that there is a relationship when it does not
exist
...
For instance, in medical
fields, where the cost of committing type 1 error can be quite costly, the p-value is placed at 0
...
There is
only 1% chance that you may state that a relationship or difference exists when it actually does not
...
05, to minimize the chances of committing
type 1 error
...
This means, stating there is
no significant relationship between variables or there is no significant difference between groups, when
in actual sense the relationship or the difference exists
...
For instance, if the analysis found no
relationship between increasing advertising budget and sales revenue, this could result to the market
researcher not adding significant investment on advertising
...
Usually differences or relationships are not spotted when samples selected are too small, the
smaller the samples, the greater the chance that any effects or relationships or differences will not be
established
...
8
...
Summary
In this section, we covered the two main techniques of analyzing data: quantitative and
qualitative analysis
...
Descriptive analysis are used to describe the sample in terms of
measures of central tendency and measures of dispersion
...
Inferences are of three types:
univariate, bivariate, and multivariate
...
108
8
...
1
...
An exploratory research, that aims at exploring issues that have not been
verified by any theory, then qualitative analysis are more appropriate
...
6
...
Activities
From the identified marketing problem, and the stated research objectives; evaluate if the research
objectives are sourcing for quantitative or qualitative data
...
Using the steps of hypothesis testing, do
the analysis to identify if there are significant relationships between variables or significant differences
between groups
...
6
...
Further reading
Aaker, D
...
and Day, S
...
(1990) Marketing Research 4th Ed
...
Quee, W
...
(1999) Marketing research 3rd Ed
...
B
...
E
...
(1998) State of the art Marketing r Research 2nd ed
...
USA
...
W
...
and Stasch S
...
7th Ed
...
USA
...
J
...
K
...
109
8
...
4
...
Discuss various ways in which qualitative data is analyzed
Compare and contrast the T-Test for independent samples and T-test for dependent samples
Provide an example of how ANOVA can be used as an analysis method
In what situations could it be more appropriate to carry out a correlation analysis compared to
a regression analysis
...
Marketing Research Report Writing
This topic concludes the market research process
...
Hence, it should be prepared in a manner that
communicates to the intended audience
...
1
...
2
...
The report is
the only documentation of the research
...
• The research report is the only aspect of the study that marketing executives are exposed to
...
• The research report will also help determine whether a particular researcher will be used in the future
or not
...
111
9
...
Types of Reports
Basically, there are two types of reports:
1
...
2
...
The writing style is designed for rapid reading and easy comprehension of the main findings of the
research
...
4
...
Consider the type of audience (readers)-The audience and not the researcher, dictates the form of
report to be written
...
The marketing manager and other managers are basically interested in the survey findings rather
than the details of research design and methodology
...
Study objectives- The survey data needs to be presented in a manner that enables the research
objectives to be accomplished
...
Objectivity-The professional researcher has the obligation of presenting the findings in an objective
manner, unprejudiced by managements’ belief or expectations
...
iv
...
Writing everything tends to overshadow the important points
...
Be concise and clear- Clarity in writing means efficient communication
...
Avoid lengthy sentences
...
vi
...
If absolutely
necessary, they can be introduced (include a brief description or explanation of such terms)
...
Inert visual aids-where necessary, visual aids such as graphs, pictures and maps can be used in the
report
...
Italics,
capitalization, exclamation marks, quotation marks and other symbols can be used appropriately to
indicate emphasis on some key points raised
...
5
...
Executive summary/Summary of findings
Abstract of information that is necessary to reach a conclusion
...
Many executives will only read the report summary, thus it must present the
major points of the results, yet be complete enough to summarize the entire document
...
Write this section last
...
For short reports (<6pages), omit this and
give distinctive headlines
...
Shows research purpose, specific research objectives, and any
hypothesis
...
The
background answers questions like-why was this study undertaken? Why was it important to study?
(Use your secondary research here, general observations, qualitative interview findings (i
...
, discussions
you’ve had with friends, etc)
...
Not only should the researcher exhibit adequate
historical knowledge, but also familiarity with
present activities and developments for the product, service, or brand and industry
...
Your task is to select pertinent facts and tie
them together for an understandable and accurate background
...
Objectives
What was the study designed to achieve? If the study had numerous objectives, state them all
...
One objective can be sub divided into minor objectives for presentation in a study
...
It is important that researchers avoid biased terms such
as to justify, or ambiguous terms such as to find out
...
g
...
The composite index or joint effect of all the specified strategies
will enable the researcher find out if the differentiation strategies have an effect on the market share
...
Methodology will outline:
What procedures were used to collect primary data (observation, questionnaire, interviews)?
How the data was collected (telephone, mail, personal interview)?
Where was the data collected?
What type of sample was used (e
...
probability and simple random sampling or nonprobability and
judgement, for example)?
How many subjects?
General characteristics of sample? (e
...
, median age; average number of flights respondents take per
year; % of respondents from Nairobi, etc
...
e
...
Survey findings
Contains analyzed data-uses tables, charts, diagrams etc
...
This section might be broken down into subsections
depending on how many objectives were identified in the introduction
...
This will involve interpreting graphs, charts, and numbers for your audience in a clear, concise,
and nontechnical manner
...
In this section, other that data presentation, it is important that important analysis are
provided
...
R-values
can also provide information on the strength and direction of the relationship
...
State any limitations regarding sampling bias, interviewer bias, time
pressures,
sample
size,
measurement
error,
etc
in
this
section
Also include in this section things you would change if you did this study again
...
Conclusions and recommendations
Make conclusions on the results, guided by the research objectives
...
Conclusion and Recommendations are derived specifically from the research findings
...
Each research objective should have a conclusion
...
114
Appendix
Serves the function of catch-all for the report
...
Thus you can have different
appendices such as:
Questionnaire
Detailed calculations (if any)
General tables (if any)
Any other support material
References
References used in the survey are indicated here
...
The difference between a reference list and a bibliography, is that a
reference list only references those articles, books or written material that is cited in the text, whereas
the bibliography is more extensive and include all written material that the researcher consulted
although are not included or referenced in the document
...
It is expected that all information presented in the
research report will be clearly documented in the body of the report as well as in this section
...
In academic
and market research, the APA style of referencing is used
...
muhlenberg
...
Examples of citating work in APA format
Book:
Strunk, W
...
, & White, E
...
(1979)
...
New York, NY:
Macmillan
...
, & Parry, T
...
Designing brain-compatible learning (3rd ed
...
Chapter of a Book:
Bergquist, J
...
(1992)
...
In J
...
Buenker & L
...
Ratner (Eds
...
53-76)
...
Journal Article with DOI:
Paivio, A
...
Perceptual comparisons through the mind's eye
...
doi:10
...
24
...
225
115
Journal Article without DOI (when DOI is not available):
Becker, L
...
, & Seligman, C
...
Welcome to the energy crisis
...
Hamfi, A
...
(1981)
...
E-journal of Applied Psychology, 2(2), 38 -48
...
lib
...
edu
...
php/fdo
Online Newspaper Articles:
Becker, E
...
Prairie farmers reap conservation's rewards
...
Retrieved from http://www
...
com
Encyclopedia Articles:
Brislin, R
...
(1984)
...
In R
...
Corsini (Ed
...
1,
pp
...
New York, NY: Wiley
...
(2005)
...
Retrieved from http://0www
...
com
...
muhlenberg
...
(2001,
March 15)
...
Retrieved from
http://www
...
com/2000/index
...
, & Zukow-Goldring, P
...
Is modeling knowing? [Review of the book Models of
cognitive development, by K
...
American Journal of Psychology, 114, 126-133
...
Data Sets:
Simmons Market Research Bureau
...
Simmons national consumer survey [Data file]
...
Blog post:
Lincoln, D
...
(2009, January 23)
...
[Web log post]
...
blogspace
...
php
Website with no author or date of publication:
Census data revisited
...
d
...
edu/data/index
...
If no DOI has been assigned to the content, provide the homepage URL
...
6
...
The research report should:
• Address the problem- it should not assume that the reader has prior knowledge of the problem
situation, but should give all relevant information
...
• Execution of the research procedures should be well laid out
...
• The report should make proper use of numbers and statistics that are not misleading to readers
...
Any assumptions made in the
interpretation should be clearly identified
...
• It should clearly specify the target population to which the findings will apply
...
9
...
Summary
At the end of the research process, the researcher needs to communicate his findings by way of a
report and/or oral presentation
...
Popular reports are meant for use by many readers, including managers, and hence should
be easy to understand and interpret
...
7
...
Note
The research report should always be written the target audience in mind
...
117
9
...
2
...
9
...
3
...
A
...
G
...
John Wiley and sons,
New York
...
T
...
Butterworth-hiennemann Oxford
Blankenship, A
...
Breen, G
...
and Dukta, A
...
Contemporary publishing group Inc
...
Boyd, H
...
Westfall, R
...
F (1990) Marketing Research texts and cases
...
Richard D Irwin Inc
...
Baker, M
...
(1991) Research for marketing, The Macmillan Press Ltd Hong Kong
...
7
...
Self-Test Questions
Discuss the components of a research report
What are the qualities of a good research report?
Who are the main stakeholders to bear in mind when developing the research report?
In what ways does a market research report differ from an academic research report?
118
Title: MARKETING RESEARCH
Description: Student will gain a deeper understanding of marketing research and its value in analyzing consumer markets and the environment.
Description: Student will gain a deeper understanding of marketing research and its value in analyzing consumer markets and the environment.