Marketing Research and Information Systems

1
Marketing Research and Information Systems

• Dr. John T. Drea
• Professor of Marketing
• Western Illinois University

2
The Marketing Research Process
Define the problem
Early identification of solution
Analyze the situation (plan research design,
specify sampling procedures)
Collect the data
Analyze the data
Solve the problem (prepare and present the
report, follow-up)
3
Identifying the problem

• Break the problem down into its most basic
  elements
• Dont prescribe the solution in how you define
  the problem.
• Ex Why dont customers like our products
  anymore?
• What is it you want to know?
• Exploratory research gather preliminary info. to
  help define the problem and suggest hypotheses.
• Descriptive research to describe marketing
  problems, situations, or market characteristics.
• Causal research to test hypotheses about
  cause-and-effect relationships

4
Analyze the situation (research design and
sampling issues)

• Situation analysis - an informal study of what
  information is available in the problem area.
  Useful for refining the problem definition.
• Be sure to talk with customers, sales reps, etc.
• Plan how to gather the data
• Primary data data specifically collected to
  solve your particular problem
• Secondary data information collected/published
  previously.
• Internal sources company records, databases,
  etc.
• External sources previous surveys, Internet,
  government, trade associations, etc.
• Be sure to assess the quality of the data!

5
Analyze the situation (research design and
sampling issues)

• Choosing a sample
• Who is to be surveyed? What member of the
  decision making unit will you survey, and how
  will you make sure you get them?
• How many people should be surveyed? Larger
  samples are more reliable, but more expensive.
• How should the people in the sample be chosen?
  (costs vs. expediency)

6
Analyze the situation (research design and
sampling issues)

• Key sample-related terms
• Population the total group you are interested in
• Sample the segment of the population selected to
  represent the population as a whole
• Probability sample every member of the
  population has a known probability of being
  included in the sample
• Random sample every member of the population has
  an equal probability of being included in the
  sample
• Non-probability sample every member of the
  population does not have an equal probability of
  being included in the sample.

7
Collect the data

• Focus groups a qualitative technique.
• Interviewing 6-10 people in a group setting on
  the problem area
• Allows you to explore and ask why questions and
  create interaction between subjects.
• A good first step in doing primary research.
• Lower in cost and fast (compared to other forms
  of data collection.)
• Usually up to 1.5 hours in length.
• The funnel technique on questions is common.
• Difficult to analyze - qualitative data.
• Are the subjects representative of the population?

8
Collect the data

• Types of questions Open-ended
• Completely unstructured
• Why did you choose Brand X toothpaste?
• Word association
• Whats the first word that comes to mind when you
  hear toothpaste?
• Sentence completion
• When I choose a toothpaste, the most important
  thing is ____________?
• Story/Picture completion

9
Collect the data

• Types of questions Closed ended
• Dichotomous (ex yes/no, gender)
• Multiple choice
• Likert scale a statement anchored by strongly
  agree/strongly disagree

Likert scale example I would recommend Brand X
to a friend. Strongly Agree 7 6 5 4 3
2 1 Strongly Disagree
10
Collect the data

• Types of questions Closed ended (continued)
• Semantic differential respondents select a point
  between pairs of bi-polar adjectives

Brand X is Good ___ ___ ___ ___ ___ Bad Positive
___ ___ ___ ___ ___ Negative Old
fashioned ___ ___ ___ ___ ___ Modern
11
Collect the data

• Types of questions Closed ended (continued)
• Rank Order Scale

Please rank the following brands in order of
their attractiveness to you (1 most attractive,
2 next most attractive, etc.)
Constant sum scale Please assign
points to each of the following brands
indicating your likelihood of purchase next
time. Be sure your total equals 100 points. ___
Brand X ___ Brand Y ___ Brand Z 100 pts.
12
Whats wrong with these questions?partial
source Kotler, Philip and Gary Armstrong (1999).
Principles of Marketing, 8th Edition. Upper
Saddle River, NJ Prentice-Hall.
1. What is your income to the nearest hundred
dollars? 2. How important are comfort and
reliability in an automobile? Very important
7 6 5 4 3 2 1 Very unimportant 3.
Do your children behave themselves in church?
___ yes ___no 4. How many soft drinks did you
drink last year? _______ 5. What are the most
salient attributes in your evaluation of a
summer camp? 6. Do you think it is right to
deprive your child of the opportunity to grow as
a mature person through the experience of
attending summer camp? 7. How would you rate
todays meal? ___excellent ___very good ___good
___unacceptable
13
Analyze the data

• Determine how you will analyze your data before
  you begin collecting data.
• Establish your coding format before collecting
  data.
• Remember, you are answering the research
  question.
• Dont be afraid to ask the so what question.
• Beware data mining
• Type I error found something when nothing was
  there
• Type II error found nothing when something was
  there

14
Solve the problem (prepare/present report,
follow-up)

• Use the data to specify a course of action.
• Be sure the course of action specified logically
  flows from the data.
• Dont cut corners here - this is why you did the
  research in the first place - TO SOLVE THE
  PROBLEM. Dont bury your audience with
  chi-squares, F-ratios, and factor loadings -
  prepare the report in terms appropriate for your
  audience.