Research Process &

Scientific Method

Dr. A. K. Dey

Research Process: Dr. Dey 1

 Learning Outcome
 Classification of research based on objectives
 Characteristics (Principles) of good research
methodology
 Scientific method: Validity & Reliability
 Research Designs – Different types of
 Techniques of data collection
 Characteristics (Principles) of good research design
 Difficulty in employing scientific method in marketing
research
 Steps of a marketing research process
 Problems in employing scientific method at each
research process step and how to minimize errors

Research Process: Dr. Dey 2

 Objectives of Research
 To gain further insight into the problem –
Exploratory or Formulative Research
 To describe the characteristics of a situation or a
group – Descriptive Research
 To determine the frequency with which some
variable occurs – Diagnostic Research
 To test a hypothesis of causal relationship
between variables – Hypothesis testing research

Research Process: Dr. Dey 3

 Characteristics of a Good Research
Methodology
 Systematic: Structured with specified steps to
be taken in predetermined sequence according
to a specific set of rules
 Logical: Should be guided by rules of logical
reasoning
 Empirical: Related to one or more aspects of
real situation
 Replicable: Results can be verified by
replicating the study

Research Process: Dr. Dey 4

 Scientific Method
 Two main characteristics of scientific
method
 Validity: The claim of measuring the identified
variable
 Reliability or Repeatability: A repeat of the study
should lead to the same outcome; like experiments
of Physics or Chemistry
 Very few Marketing Research projects could
qualify as experiments; much less
reproducible experiments
Research Process: Dr. Dey 5
 What is Research Design?
 A plan for selecting the sources and types
of information used to answer research
questions
 A framework for specifying the
relationships among the study variables
 A blueprint that outlines each procedure
from the hypothesis to the analysis

Research Process: Dr. Dey 6

 Descriptors of Research Design
1. Degree of crystallization of research question
2. Method of Data Collection
3. Power to produce effects
4. Purpose of the study
5. Time dimension
6. Topical scope – breadth & depth
7. Research environment
8. Participants’ perceptions

Research Process: Dr. Dey 7

 1. Degree of crystallization of Research
question
 Exploratory study is usually to develop
hypotheses or questions for further
research

 Formal study is to test the hypotheses or

answer the research questions posed

Research Process: Dr. Dey 8

 2. Methods of Data Collection
 Monitoring, which includes observational
studies

 Interrogation/communication studies

Research Process: Dr. Dey 9

 3. Power to Produce Effects
 In an experiment, the researcher attempts
to control and/or manipulate the variables
in the study

 In an ex post facto design, the researcher

has no control over the variables; they can
only report what has happened

Research Process: Dr. Dey 10

 4. Purpose of the Study
 Exploratory Studies Tries to define the
problem more objectively

 Descriptive study tries to explain

relationships among variables

 Causal study is how one variable

produces changes in another
Research Process: Dr. Dey 11
 5. The Time Dimension
 Cross-sectional studies are carried out
once and represent a snapshot of one
point in time

 Longitudinal studies are repeated over an

extended period

Research Process: Dr. Dey 12

 6. The Topical Scope
 Statistical studies attempt to capture a
population’s characteristics by making
inferences from a sample’s characteristics

 Case studies place more emphasis on a

full contextual analysis of fewer events or
conditions and their interrelations

Research Process: Dr. Dey 13

 7. The Research Environment
 Field conditions: Actual environmental conditions

 Laboratory conditions: Manipulated conditions

 Simulations: Major characteristics of various

conditions and relationships in actual situations
are often represented in mathematical models:
Role plays, Mystery shoppers

Research Process: Dr. Dey 14

 8. A Participant’s Perceptions
 Usefulness of a design may be reduced when
people in a disguised study perceive that
research is being conducted

 Participants’ perceptions influence the outcomes

of the research: Example if the sales girl comes
to know that her performance is being watched
by mystery shoppers, the performance itself may
change

Research Process: Dr. Dey 15

 Why do Exploratory Studies?
 Exploration is particularly useful when
researchers lack a clear idea of the
problems
 Outcome of an Exploratory study is a set
of Hypotheses

Research Process: Dr. Dey 16

 Data Collection Techniques
 Two types: Qualitative & Quantitative
 Qualitative techniques: Suitable for exploratory
studies
 Secondary data
 Experience surveys
 In-depth interviews
 Focus groups
 Two-stage design

Research Process: Dr. Dey 17

 Data Collection Techniques
 Quantitative techniques: Suitable for descriptive
and causal studies
 Only two types: ‘Ask’ or ‘Observe’
 Observation
 Survey
 Experimentation

Research Process: Dr. Dey 18

 Causation
 The essential element of causation is
 A “produces” B

or
 A “forces” B to occur

Research Process: Dr. Dey 19

 Causal Study Relationships
 Possible relationship between two variables

 Symmetrical
 Variables fluctuate together but the changes in neither
variable are due to the changes in the other

 Reciprocal
 Variables mutually influence or reinforce each other

 Asymmetrical
 Changes in one variable (independent) are responsible for
the changes in the other (Dependent)

Research Process: Dr. Dey 20

 Asymmetrical Relationships
 Four types of Asymmetrical Causal Relationship
 Stimulus-Response
 A change in work rules leads to higher worker output

 Property-Disposition
 Age and attitudes about saving, Social class and opinion about
taxation

 Disposition-Behavior
 Job satisfaction & work output

 Property-Behavior
 Age and participation in sports
Research Process: Dr. Dey 21
 Principles of Experimental Design
 Main objective is to achieve
 Validity
 Reliability

 Principles are
 Randomization
 When ever the researcher faces a situation to exercise choice,
random method must be used

 Replication
 Design should have built-in replications so as to compare outcomes

 Local Control
 Using best available method instruments
Research Process: Dr. Dey 22
 MR & Scientific Method
 It is very difficult to measure identified
variable: Example – TV viewer ship
 Controlling variables
 Highly interactive; almost impossible to control
in a market place
 Projects mostly carried out “one time” – hence
no way to test reliability
 MR projects are proprietary – hence no way to
check reproducibility
Research Process: Dr. Dey 23
Scientific Vs Non – Scientific Methods
 Major differences between Scientific &
Non Scientific Method that affect Validity &
Reliability of the results are
 Objectivity of the investigator
 Accuracy of measurements
 Degree to which the investigation is
continuing & exhaustive

Research Process: Dr. Dey 24

Scientific Vs Non – Scientific Methods
 Objectivity of the investigator
 A researcher must base his judgment on facts
& not on preconceived notion or intuition
 Two major factors that reduce the objectivity
on the part of the investigator
 MarketResearchers report to a strong willed
executive who thinks that he knows the market
well
 Marketers often tend to exploit comfortable
segments; They ignore segments that are different,
unattractive or threatening
Research Process: Dr. Dey 25
Scientific Vs Non – Scientific Methods
 Accuracy of measurements
 In experiments of Physics or Chemistry measuring
devices of great accuracy are used
 In MR attitudes, intensions, behaviours etc are
measured – so far no instruments can measure these
accurately
 Such variables are dynamic – checking reproducibility
is not possible
 Even crude devices like Attitude Rating Scales are
administered by different interviewers & interviews are
conducted in different settings – results not
reproducible
Research Process: Dr. Dey 26
Scientific Vs Non – Scientific Methods
 Continuing & Exhaustive nature of
investigation
 Scientists are continuously & aggressively
searching for additional evidence, they are not
sure that ultimate truth has been found
 Marketing Research tends to be less
continuous than science
 Results of MR projects not published or
shared – no refinement done
Research Process: Dr. Dey 27
 Difficulties in Applying Scientific
Method to Marketing Research
 Investigator involved in use of results
 Imprecise measuring devices
 Influence of measurement process on the results
 Time pressure for results
 Difficulty in using experiments to Test
Hypothesis
 Great complexity of the subject : Complexity of
human behaviourial response makes it difficult to
achieve high level of Validity & Reliability
Research Process: Dr. Dey 28
 Research Process
 Seven inter-related steps
1. Specifying research objectives
2. Preparing a list of needed information
3. Designing the data collection project
4. Selecting a sample type
5. Determining sample size
6. Organizing & carrying out the field work
7. Analyzing the collected data & report the
findings

Research Process: Dr. Dey 29

 Research Process & Problems in
Achieving Scientific Method
 To achieve Validity & Reliability,
Marketing Research should be
conducted as a proper Scientific Method
 At each of the seven steps let us analyze
A. Problems in achieving Scientific Method
B. Steps to minimize the potential sources of
errors

Research Process: Dr. Dey 30

 1. Specifying Research Objectives
A. Problems in achieving Scientific Method
a. Manager’s Expectations of research results
B. Minimizing potential sources of errors
a. Write research objectives
b. Manager & researcher must discuss the
objective statements & if necessary modify

Research Process: Dr. Dey 31

 2. Preparing a list of needed
information
 Problems in achieving Scientific Method
a. Due to busy schedule manager may not get
adequately involved
b. May think researcher knows what to do
 Minimizing potential sources of errors
a. Manager & researcher should develop ‘List of
needed information’ together & evaluate usefulness
Research is not needed if manager is forced to select a
particular course of action irrespective of research
findings
Research Process: Dr. Dey 32
3. Designing data collection project
 Problems in achieving Scientific Method
a. Using inappropriate research design
b. Wrong selection of respondents
c. Asking unclear or ambiguous questions
d. Using large scale study instead of small
scale & vice versa
e. Using poor experimental design

Research Process: Dr. Dey 33

3. Designing data collection project
 Minimizing potential sources of errors
Five important issues that must be addressed:
 Should the research be exploratory or
conclusive?
 Who should be interviewed & how?
 Should only few cases be studied or large
samples?
 How well experiments be incorporated?
 How should data collection form be
designed?
Research Process: Dr. Dey 34
 4. Selecting a sample type
Why samples? Probability Vs. Non
Probability
 Problems in achieving Scientific Method
a. Sample not representative of the population
 Minimizing potential sources of errors
a. Define sampling frame carefully
b. Select proper sampling method – Simple
Random

Research Process: Dr. Dey 35

 5. Determining Sample size
 Problems in achieving Scientific Method
a. Sample size depends upon
i. Nature of the problem
ii. Budget
iii. Accuracy needed
b. Small sample – Lower reliability
Large sample – Likely to give higher reliability
 Minimizing potential sources of errors
a. Use Sampling Statistics to calculate sample size for
a given accuracy (Confidence Interval)
b. Care exercised in determining sample size &
sample type will minimize errors
Research Process: Dr. Dey 36
 6. Organizing &
Carrying out field work
 Field work: Selecting, Training, Controlling & Evaluating
field force
 Involves substantial portion of budget
 Potential source of errors through lack of Validity &
Reliability
 Problems in achieving Scientific Method
a. Varying skills of field workers
b. Forms filled without interview
c. May not follow instruction
d. Investigator’s bias
e. Respondents’ bias
Research Process: Dr. Dey 37
 6. Organizing &
Carrying out field work
 Minimizing potential sources of errors
a. Follow good practices in selection, training,
controlling & evaluating field workers
b. Incorporate Back Checks & Spot Checks
c. Motivate supervisors
d. Deploy adequate field force – release time
pressure

Research Process: Dr. Dey 38

 7. Analyzing Data &
Report Preparation
 Problems in achieving Scientific Method
a. Care & precaution not taken during editing,
coding & data entry
b. List of needed information not prepared properly
c. Research objectives not established correctly
 Minimizing potential sources of errors
a. Editing& Coding done carefully
b. Incorporate extensive validity checks
c. Inferences to be drawn based on factual data &
not based upon researcher’s personal
understanding
Research Process: Dr. Dey 39