The Marketing Research Process Chapter 3 Audhesh Paswan, Ph.D.

advertisement
The Marketing Research Process
Chapter 3
Audhesh Paswan, Ph.D.
INFORMATION


REDUCES UNCERTAINTY
HELPS FOCUS DECISION MAKING
Marketing Research can be . . .
Accurate
Fast
Inexpensive
Pick two! Can’t have three at the same time!
STAGES IN THE RESEARCH
PROCESS






PROBLEM DISCOVERY AND PROBLEM
DEFINITION
RESEARCH DESIGN
SAMPLING
DATA GATHERING
DATA PROCESSING AND ANALYSIS
CONCLUSIONS AND REPORT
1


Establishing the need for
marketing research.
Is it needed?
May not be needed:
– information may already be available
– not enough time to do study
– not enough money
– costs may outweigh value of research
Marketing Research

Marketing Research Types
Problem Identification
Problem-Solving
* Market Potential Research
* Market Share Research
* Image Research
* Market Characteristics
* Sales Analyses Research
* Forecasting Research
* Business Trend Research
* Segmentation Research
* Product Research
* Pricing Research
* Promotion Research
* Distribution Research
2



Define the problem
Most important part - everything else is based
upon this!
May do “exploratory research” to help define
the problem
Think of yourself as a “marketing doctor”
– make sure you can tell the symptoms from
the problem..
 specify
the symptoms > itemizing the possible
causes of the symptoms > listing the reasonable
alternative course of action.
Defining the Problem Results in
Clear Cut Research Objectives
Symptom Detection
Analysis of the Situation
Exploratory Research
(Optional)
Problem Definition
Statement of Research Objectives
“The formulation of the problem is
often more essential than its
solution”
Albert Einstein
The Process of Problem Definition
Ascertain
the
decision
maker’s
objectives.
Understand
the
background
of the
problem.
Isolate and
identify the
problems,
not the
symptoms.
State the
research
questions
and
research
objectives.
Determine
the unit
of analysis
Determine
the
relevant
variables
3


Establish research
objectives.
What information is needed to solve the
problem?
Set objectives associated with this information.
I keep six honest serving men,
(they taught me all I knew), their
names are what, and why, and
when, and how, and where and
who.”
--Rudyard Kipling
4


Determine research design
Exploratory Research
– unstructured, informal, and sometimes
intuitive
Descriptive Research
– very common in marketing research
– descriptive in nature
– involves communication and/or observation
for data collection
– lends itself to statistical analysis
4
Determine research design
Causal Research
establish cause and effect relationship
problem: multiple causes and effects
problem: hard to isolate
involves experiments
e.g., pretest, posttest, control groups
Education
Income
Happiness
4
Research Design. .
Exploratory
Descriptive
Causal
DEGREE OF PROBLEM DEFINITION
Exploratory Research
(Unaware of Problem)
“Our sales are declining and
we don’t know why.”
“Would people be interested
in our new product idea?”
Descriptive Research
(Aware of Problem)
Causal Research
(Problem Clearly Defined)
“What kind of people are buying “Will buyers purchase more of
our product? Who buys our
our products in a new package?
competitor’s product?”
“Which of two advertising
“What features do buyers prefer campaigns is more effective?”
in our product?”
4

Research Design - I
Exploratory Research:
Objective:
Discovery of ideas and insights.
Characteristics: Flexible, Versatile, Unstructured,
Often the Front End of total Research Design,
Small Non-representative Sample,
Analyses typically qualitative.
Findings:
Tentative, typically followed by further
exploratory, descriptive or causal research.
Methods:
Literature Search, Focus Groups, Experience
Surveys, Pilot Surveys, Expert Interviews, Case
Studies, Reliance on Secondary Data.
4

Research Design - II
Descriptive Research:
Objective:
Describe Market Characteristics or Functions,
Test Specific Hypotheses.
Characteristics: Prior Formulation of Hypotheses,
Preplanned, Formal and Structured Design,
Information needed is predefined,
Sample is Large and Representative,
Data Analyses typically Quantitative.
Findings:
Making.
Conclusive, used as input into Decision
Methods:
Surveys, Panels and Observation
(Typically Primary Data).
4

Research Design - III
Causal Research:
Objective:
Determine Cause and Effect Relationship,
Test Specific Hypotheses.
Characteristics: Manipulation of Independent Variables, and
Control of Other Mediating Variables.
Prior Formulation of Hypotheses, Preplanned,
Formal and Structured Design, Information
needed is predefined, Sample Representative,
Data Analyses typically Quantitative.
Findings:
Making.
Conclusive, used as input into Decision
Methods:
Experiments (Typically Primary Data)
IDENTIFYING CAUSALITY
A causal relationship is impossible to prove.
Evidence of causality:
1. The appropriate causal order of events
2. Concomitant variation--two phenomena
vary together
3. An absence of alternative plausible
explanations
If you do not know
where you are going,
any road will take you there.
RESEARCH DESIGN



MASTER PLAN
FRAMEWORK FOR ACTION
SPECIFIES METHODS AND PROCEDURES
5

Identify information types
and sources
Two types of information:
– Secondary data
 information
already collected for some other
purpose
 internal or external
 typically used in exploratory research
 some key problems??
– Primary data
 information
collected to specifically answer the
problem
 observation or communication method
5


Information - Data
Data - Known facts or things used as basis for
inference; information; material to be
processed and stored.
Information - what is told, knowledge, news,
charge or accusation.
Data
Information
5

Information Sources
Four major sources.
Intuition
Authority
Decision Making
Process
Research
Experience
5



Marketing Research Data
Secondary vs Primary
Qualitative vs Quantitative
Internal vs External
Secondary
Primary
Anecdotes
Experience
Case studies
Opinions, etc.
Focus groups
Interviews
projection
techniques, etc.
Census
Syndicated data
Journals
Magazines, etc.
Surveys
Observations
Experiments
Tests, etc.
Qualitative
Quantitative
5
Marketing Research Data
 Marketing
Research Data
Secondary
Data
Qualitative Data
Descriptive
Survey data
Observation
Primary data
Quantitative data
Causal
Experiment
6


Determine methods of
accessing data
Depends on what kind of data is needed
Methods different for secondary data collection
than for primary data collection, e.g.,
– Secondary data - library, internet, buy
syndicated data, CD-ROM, etc.
– Primary data - mail, telephone, mall
intercept, door-to-door, etc.
Data Collection

Qualitative Research
Objective:
To gain understanding of the underlying
reasons and motives (Exploratory stage).
Sample:
Small number, nonrepresentative.
Method:
Unstructured.
Analyses:
Nonstatistical.
Outcome:
Develop initial understanding.
Data Collection

Quantitative Research
Objective:
To quantify the data, and generalize the
results to the population of interest.
Sample:
Large numbers, Representative.
Method:
Structured.
Analyses:
Statistical.
Outcome:
Recommend a final course of action.
Data Collection - Methods

Direct
Indirect
(Nondisguised)
(Disguised)
Focus
Groups
Association
Qualitative Data
Depth
Interviews
Completio
n
Construction
Projective
Techniques
Expressive
Data Collection - Methods

Quantitative/ Primary Data
Communication
Observation
Versatility
Speed
Cost
Objectivity
Accuracy
Relevant for:
Demographics, Socioeconomic, Psychological/Lifestyle
Characteristics;
Attitudes, Opinions, Awareness, Knowledge, Intentions, Motivations,
and Behavior.
Data Collection - Methods

Communication or Surveys
Telephone
In Home
Traditional
Telephone
CATI
Personal
Mall Intercept
Mail
Interview
Mail
CAPI
Mail panel
Data Collection Methods - Comparison
Criteria
Flexibility of data collection M
Diversity of Questions
Use of physical stimuli
Quantity of data
Response rate
Speed
Cost
Interviewer bias
Sample control
Field force control
Sensitive Information
L=Low, H=High, M=Medium.
Telephone
H
L
L
L
M
H
M
M
M/H
M
H
Personal
L
H
H
H/M
H
M/H
M/H
H
M/H
L/M
L
Mail
M
M
M/H
L/M
L
L
No
L/M
H
H
Data Collection - Methods

Observation
Personal
Observation
Mechanical
Observation
Audit
Content
Analyses
Trace
Analyses
Data Collection - Methods

Experiments - Test Marketing.

Simulated, Controlled, Standard, and National Rollout.
Factor
Laboratory
Environment
Realistic
Control
Reactive Error
Demand Artifact
Internal Validity
External Validity
Time
Number of Units
Implementation Ease
Cost
Artificial
High
High
High
High
Low
Short
Small
High
Low
Field
Low
Low
Low
Low
High
Long
Large
Low
High
7






Design data collection
forms
Depends on type of research being conducted
Some key issues:
Structured or unstructured
Disguised or undisguised
Number of questions.
Wording and sequencing of questions.
Types of Variables
Categorical
Continuous
Dependent
Independent
Measurement Instrument

Communication methods typically use a
questionnaire as the instrument.

The questionnaire must motivate the
respondents to cooperate, become involved, and
provide complete and accurate answers.
Measurement Instrument
1.
2.
3.
4.
Specify the information needed
Type of interviewing method
Content of individual questions
Design the questions to overcome inability
and unwillingness
5. Decide on the question structure
6. Determine the question wording
7. Arrange the questions in proper order
8. Identify the form and layout
9. Reproduce the questionnaire
10. Eliminate bugs by pretesting.
Measurement Instruments

Marketers want to measure:
Demographics/Socioeconomic Characteristics
Psychographics and Lifestyles
Personality
Motivation
Consumer knowledge regarding
Product - Awareness, Attribute and Price.
Purchase - Where and When of purchase.
Usage - Usage operations and situations.
Past Behavior,
Attitudes and Opinions,
Behavioral Intentions, etc.
Measurement Scales
1. Nominal - identify and classify
(Sex, user-nonuser, etc.; Descriptive - percentage and mode; Inferential
- Chi-square, binomial tests)
2. Ordinal - relative position but not magnitude of
difference
(Quality/ preference ranking, market position, etc.; Descriptive - %,
median; Inferential - Rank-order Correlation, Friedman ANOVA)
Measurement Scales
3. Interval - differences, arbitrary zero point.
(Temp, attitudes, opinions, index numbers, etc.; Descriptive - range,
mean, standard deviation; Inferential - Correlation, t-tests, ANOVA,
regression, and multivariate analyses).
4. Ratio - fixed zero point, ratios.
(Length, weight, age, income, sales, market share, etc.; Descriptive geometric & harmonic mean; Inferential - Coefficient of variation)
Measurement Scales



Attitudes, Opinions , Preferences and
Perceptions.
“When you can measure what you are
speaking about and express it in numbers, you
know something about it.” Lord Kelvin.
Scales should be evaluated for reliability and
validity.
Measurement Scales
Operationalization of scales for measuring Attitudes, Opinions ,
Preferences and Perceptions.
 Usually an adaptation of Interval scale.
1. Continuous rating scale - mark on a continuous line.
2. Itemized Rating Scale, e.g.,
Likert Scale - five point (Strongly agree to Strongly disagree) scale.
Semantic differential scale - seven point scales with bipolar labels.
Staple scale - Unipolar ten-point scale, -5 to +5, without a neutral

8




Determine sample plan
and size
Who are you going to sample -respondent?
How many are you going to sample - sample
size?
Depends upon
– Time
– Money
– Response rate
– Type of data collection form
Some key terms - sample elements, sample frame,
sampling plans, sample size.
SAMPLING




SUBSET OF POPULATION
WHO IS TO BE SAMPLED
HOW LARGE A SAMPLE
HOW WILL SAMPLE UNITS BE SELECTED
Sampling

Sampling Techniques:
1. Nonprobability:
Convenience, Judgmental, Quota, and
Snowball Sampling.
2. Probability:
Simple, Systematic, Stratified (Proportionate,
Disproportionate), Cluster (One-stage, Two-stage),
and Others.
9




Collect data
Trained interviewers
Questionnaires
Data collection companies
Avoid non-sampling errors
10





Analyze data
Give meaning to raw data - interpretation
Involves - data cleaning, coding, tabulation, crosstabulation, statistical tests, & interpretation.
Descriptive statistics - frequencies, mean, median, mode,
SD, etc.
Statistical analysis
– tests of association - cross tabs, correlation,
regression, etc.
– Test of difference - t-test, f-test, ANOVA etc.
Presentation
HYPOTHESIS



AN UNPROVEN PROPOSITION
A POSSIBLE SOLUTION TO A PROBLEM
GUESS
Data Analyses








Preliminary data analyses plan.
Questionnaire checking.
Editing (field and office).
Coding.
Transcribing.
Data cleaning.
Statistical adjustments.
Selecting a data analyses strategy.
Data Analyses

Descriptive statistics:
1. Frequencies (simple count).
2. Mean (average), median (50% point),
mode (most frequent occurrence).
3. Standard deviation (Sq. root of variance),
variance (mean squared deviation from).
4. Skewness (quirks in data).
Data Analyses

Two basic tests of relationships:
1. Tests of Association.
e.g., is purchase behavior related with (or dependent on)
income or gender.
2. Tests of Differences.
e.g., do men differ from women in the way they buy
food, cloths, shoes, or
the way they view life, or
their readership or viewership habits etc.
Data Analyses

Tests of Association.
Cross Tabulation (Chi Square),
Correlations and Regressions,
ANOVA etc.
Data Analyses

Tests of Differences.
T-tests, ANOVA, Discriminant etc.
Data Analyses

Advanced Multivariate Techniques.
Some examples:
ANOVA, MANOVA, Multiple Regressions;
Factor Analyses, Cluster Analyses,
Discriminant Analyses;
Multidimensional Scaling;
Conjoint Analysis;
Structural (Causal) Modeling.
11




Prepare and present final
research report
Communicate study results to client
Importance cannot be underestimated!
Determine exactly how information should be
presented - oral versus written.
Some tools for presenting data and results:
– frequency tables
– charts and graphs
– mean tables, etc.
RESEARCH PROPOSAL


A WRITTEN STATEMENT OF THE
RESEARCH DESIGN THAT INCLUDES A
STATEMENT EXPLAINING THE PURPOSE
OF THE STUDY.
DETAILED OUTLINE OF PROCEDURES
ASSOCIATED WITH A PARTICULAR
METHODOLOGY
Download