Marketing Research Process

advertisement
CHAPTER 2
Problem Definition and The
Marketing Research
Process
The Marketing Research Process
5. Data collection
How?
What?
Why?
4. Selecting the
Sampling procedure
6. Data analysis
3. Choosing the
research method
2. Creating the
research design
1. Identifying &
Formulating the
Research Problem
8. Follow-up
7. Report writing &
presentation
Overview of the Research
Process (alt.)
1. Establish need for information
2. detail research objectives and information needs –
“Why is this project being conducted?” and "What specific information
is required to attain the objectives?"
3. design research and data sources – information to collect, sources of
data, and plan guiding data collection and analysis
4. design data-collection procedure – e.g., questionnaire
5. design sample – define population, sample-selection method, and the
sample size
6. collect data
7. process and code data
8. analyze data
9. present results – oral presentation and/or written report
Step 1: Define the problem
 Objectives Vs. problems
 Management decision problem vs. marketing research
problem?
 Problem or Symptom?
THE CASE OF CHERRY BLOSSOM
 Cherry Blossom sales are dropping.
 Kiwi is the main competitor.
 Salespeople tell me that Kiwi is gaining share.
 LET’S KILL KIWI!!!
 what is likely to happen with such decisions?
$20 million ad campaign
Price promotions
Six months later - NO JOB!!
How to define the decision problem?
 DECION MAKING ISSUES:
 Cherry Blossom shoe polish is losing sales!
 BEWARE OF FRAMING
 Cherry’s market share is dropping….because Kiwi is
taking away market share
 WHAT ELSE COULD BE HAPPENING?
ANALYSIS. ANALYSIS. ANALYSIS!
 Is the market shrinking?
 Are there other new competitors?
 Are there different needs?
Problem Definition: Cherry Blossom Shoe Polish is losing
sales due to three important trends:
-consumers are wearing other kinds of non-leather shoe
-attitudes towards shoe polishing is declining
-consumers have less time to polish their shoes
Identifying & Formulating the
Research Problem
Packaging
Identifying & Formulating the
Research Problem
Creating the Research Design
 Research Design
 Exploratory Study (Montblanc)
 Seeking insight into the general nature of the problem.
 Descriptive Study
 Description of the environment.
 Causal Study
 Whether one variable causes the other.
 Hypotheses
 A conjecture; a possible answer to the research question
Conducting Exploratory Research
 What is Exploratory Research?
 Seeking
insight into the general nature of the
problem
 Qualitative Research; Observation, Mystery
shopping, Focus groups, In-depth interviewing,
Garbology.
NO hypothesis or vague hypothesis
Descriptive Research (Example)
 Describe the characteristics of relevant groups
 Internet users
 Who buys our products?
 Where do they buy it?
Causal Studies
 Variable
 Dependent variable
 Independent variable
 Three issues for causal studies
 Temporal sequence. Attitude  Behavior.
 Concomitant variation
 Spurious association
What is this type of research?
 Who are our users?
 Will an increase in service staff help profitability?
 What is the demographic profile of our target market?
What is this type of research?
 What are the reasons for falling sales?
 Will a change in price improve loyalty?
Choosing a Method of Research
 Survey
 Mail
 Telephone
 Intercept
 Observation
 Monitors respondents actions without direct interaction
 Experiments
Sampling Procedure
 Probability vs. Non-probability sampling
 Population Vs. Sample
 www.cnn.com
 The other steps……….
Regardless of which presidential
candidate you now support, who do
you think won the debate?
NEWSWEEK Poll




George W. Bush
19%
John Kerry
61%
No winner/about even 16%
Don't know
4%
 The NEWSWEEK poll was conducted by Princeton
Survey Research Associates, which interviewed 1,013
registered voters by telephone on Sept. 30-Oct. 2. The
margin of error is plus or minus 4 percentage points.
2-5 Errors in Marketing Research
Two main types of errors in marketing research:
 sampling errors
 unavoidable inaccuracy stemming from using samples to
estimate quantities in the population
 equals difference between the sample value and the true
underlying population value
 non-sampling errors – errors that occur in the research
process over and above the sampling error
 inadvertent mistakes
 deliberate deceptions
2-5c Types of Non-Sampling Errors
 faulty problem definition – solving the wrong problem
 incorrect population definition – wrong sample
 sampling frame not representative of the population
 e.g., phone book under-represents stock buyers
 non-response errors – refusals and not at home
 poor questionnaire design
 e.g., biased or misleading question
 measurement error – scale doesn’t match construct of interest
 improper causal inferences – assigning cause inappropriately
Download