Transcript Marketing Research Essentials 5th ed

Primary Data
Collection:
Experimentation
Chapter
Seven
Copyright © 2006
John Wiley & Sons, Inc.
Learning Objectives
1. To understand the nature of experiments.
2. To gain insight into requirements for proving
causation.
3. To learn about the experimental setting.
4. To examine experimental validity.
5. To learn the limitations of experimentation in
marketing research.
6. To compare types of experimental designs.
7. To gain insight into test marketing.
John Wiley & Son, Inc
2
What is an Experiment?
To understand the nature
of experiments.
An Experiment:
The researcher changes an explanatory, independent,
or experimental variable to observe changes in the
dependent variable.
Experimental
variable
John Wiley & Son, Inc
Dependent
variable
price
total sales
advertising
market share
3
Demonstrating Causation
To gain insight into requirements
for proving causation
• Causal Research:
– The only types of research that has the potential
to demonstrate that a change in one variable
causes some predictable change in another
variable
• To Demonstrate Causation:
– Concomitant Variation (Correlation
– Appropriate Time Order Of Occurrence
– Elimination Of Other Possible Causal Factors
John Wiley & Son, Inc
4
Demonstrating Causation
•
To gain insight into requirements
for proving causation
Scientific Definition—Causation and Causality
1. X is only one of a number of determining conditions
that caused the observed change in Y.
2. X can be a cause of Y if the presence of X makes the
occurrence of Y more probable or likely
3. One can never definitively prove that X is a cause of Y
but only infer that a relationship exists
4. Causal relationships are always inferred and never
demonstrated conclusively beyond a shadow of a
doubt.
John Wiley & Son, Inc
5
Demonstrating Causation
•
To gain insight into requirements
for proving causation
Concomitant Variation:
– A predictable statistical relationship between two
variables—A caused a particular change in B
•
Appropriate Time Order of Occurrence:
– A change in an independent variable must occur
before a change in the dependent variable
•
Elimination of Other Possible Causal Factors
– The change in B was not caused by some factor other
than A
John Wiley & Son, Inc
6
The Experimental Setting
To learn about experimental setting
• Laboratory experiments:
– Conducted in a controlled setting.
– Advantages of laboratory experiments:
• Ability to control all variables
• Greater internal validity
– Disadvantages of laboratory experiments:
• External validity
• Not transferable to the actual marketplace
John Wiley & Son, Inc
7
The Experimental Setting
To learn about experimental setting
• Field experiments:
– Tests conducted outside the laboratory
– Advantages
• Creates realism of the environment
– Disadvantages
• Internal validity
• No control over spurious factors
– Action of competitors
– Weather
– economy
John Wiley & Son, Inc
- societal trends
- political climate
8
Experimental Validity
To examine experimental validity
• Validity
– Degree to which an experiment actually measures
what it is trying to measure
• Internal and External Validity
– Internal Validity:
• The extent to which competing explanations can be
ruled out.
– External Validity:
• The extent to which causal relationships can be
generalized to outside persons, settings, and times
John Wiley & Son, Inc
9
Experimental Notation
To understand the nature of
experiments
• Experimental Treatment
– Factors whose effects are to be measured and
compared
• Experimental Notation:
–
–
–
–
X indicates exposure
O for observation
O1 X O2 for time periods
X1 O1 for simultaneous exposure and measurement
X2 O2
John Wiley & Son, Inc
10
Extraneous Variables
To examine experimental validity
• Extraneous Variables: Threats to Experimental
Validity
–
–
–
–
History
Maturation
Instrument Variation
Selection Bias
• Randomization or matching
– Mortality
– Testing Effect
– Regression to the Mean
John Wiley & Son, Inc
11
Extraneous Variables
•
•
To examine experimental validity
Confounding Variables—extraneous causal factors
because they confound the treatment condition
Four Basic Approaches to Control Extraneous
Factors
–
–
–
–
Randomization
Physical Control
Design Control
Statistical Control
John Wiley & Son, Inc
12
Experimental Design,
Treatment, and Effects
To learn about experimental setting
• Experimental Design
– Ex Post Facto Research—non-experimental
designs
• Factors of the Experimental Design:
– Researcher has control of variables and
manipulates them.
•
•
•
•
Treatment
Subjects
Dependent variable
Plan for dealing with extraneous causal factors
John Wiley & Son, Inc
13
Experimental Design,
Treatment, and Effects
•
Treatment Variable
–
•
Independent variable that is manipulated
Experimental Groups
–
–
•
To learn about experimental setting
Control Group
Test Group
--Experimental Effects
–
The effect of the treatment variable on the
dependent variable
John Wiley & Son, Inc
14
Limitations of
Experimental Research
To learn the limitations of
experimentation in marketing research
• The High Cost of Experiments
• Time
• Money
• Security Issues
• exposing to the actual marketplace
• competitors
• Implementation Problems
• cooperation
• contamination
• lack of control group
John Wiley & Son, Inc
15
Selected Experimental
Designs
To compare experimental designs
• Three Pre-experimental Designs
– Offer little or no control over extraneous factors
• One-Shot Case Study
–X
O1
– Basic weakness
• No pretest observations
• No control group of test units
• Lacks internal validity
John Wiley & Son, Inc
16
Selected Experimental
Designs
To compare experimental designs
• One-Group Pretest-Posttest Design
– O1 X O2
– Pre- and postmeasurements but no control group
– Threats
• History
• Maturation
• Only one pretest observation
John Wiley & Son, Inc
17
Selected Experimental
Designs
To compare experimental designs
• Static-Group Comparison Design
Experimental Group: X O1
Control Group
O2
– Pre-experimental with a control group
– Threats
• Absence of pretests
• Not assigned to the groups randomly
John Wiley & Son, Inc
18
Selected Experimental
Designs
To compare experimental designs
• True Experimental Designs
– Before and After with Control Group Design
• Experimental Group (R) O1
• Control Group
(R) O3
X
O2
O4
– Threats
• Mortality
• History
John Wiley & Son, Inc
19
Selected Experimental
Designs
To compare experimental designs
– Solomon Four-Group Design
•
•
•
•
•
•
Second experimental group no pretest
Second control group only posttest measurement
Experimental Group 1: (R) O1
X
O2
Control Group 1:
(R) O3
O4
Experimental Group 2:
(R)
X
O5
Control Group 2:
(R)
O6
John Wiley & Son, Inc
20
Selected Experimental
Designs
To compare experimental designs
– After-Only with Control Group
• Involves random assignment of test units to
experimental and control groups, no premeasurement of
the dependent variable.
• Experimental Group: (R) X O1
• Control Group:
(R)
O2
John Wiley & Son, Inc
21
Selected Experimental
Designs
To compare experimental designs
• Quasi-Experiments
– Developed to deal with external validity
– Researcher lacks complete over the scheduling of
treatment or must assign treatment in a
nonrandom manner.
• Interrupted Time-Series Designs
– The treatment interrupts repeated measurements.
– Cannot control history
– O1 O2 O3 O4 X O5 O6 O7 O8
John Wiley & Son, Inc
22
Selected Experimental
Designs
To compare experimental designs
• Multiple Time-Series Designs
– An interrupted time-series design with a control
group
– Experimental Group
O1 O2 O3 X O4 O5 O6
– Control Group:
O1 O2 O3
O4 O5 O6
John Wiley & Son, Inc
23
Test Markets
To gain insight into test marketing
• Test Markets Defined
– Any research that involves:
• Testing a new product or change in an existing marketing strategy.
• The use of experimental or quasi - experimental procedures
• Test Market Usage and Objectives
–
–
–
–
Estimate of market share
Effect on sales of similar products
Characteristics of consumers
Behavior of competitors
• Simulated Test Markets
John Wiley & Son, Inc
24
Test Markets
• Direct Costs of Test
Marketing
• Costs include:
–
–
–
–
–
–
–
–
Commercials
Advertising
Media
Research
POP
Coupons
Sampling
trade allowances
John Wiley & Son, Inc
To gain insight into test marketing
• Indirect Costs of Test
Marketing
• Costs include:
– Management time
– Diversion of sales activity
– Negative impact of test
market failure
– Negative trade reactions to
products
– Cost of letting competitors
know what the firm is doing
25
Test Markets
To gain insight into test marketing
• Decision To Conduct Test Marketing
• Benefits
– Good estimate of product’s sales potential
– Identify weaknesses of the product and the proposed
marketing strategy
• Factors to Consider
– Weight the cost and risk of failure
– Likelihood and sped product can be copied
– Damage an unsuccessful new product would inflict
John Wiley & Son, Inc
26
Test Markets
•
To gain insight into test marketing
Steps in a Test Market Study
1. Define the Objective
2. Select a Basic Approach
1. Simulated test market
2. Standard, or traditional, test market
3. Controlled test market
3. Develop Detailed Procedures for the Test
John Wiley & Son, Inc
27
Test Markets
4.
Select Markets for the Test
1. Minimum of 2 test sites
2. Geographically dispersed
3. Demographically
representative of US
4. Run at least 6 months
5. Media 50% of current
levels
6. Variety of media outlets
7. TV spill-over less 15%
8. TV spill-in less 15%
9. Dominant newspaper
10. Market should be at least
0.2% of US no more 2%
John Wiley & Son, Inc
To gain insight into test marketing
5.
6.
Execute the Plan
Analyze the Test Results
1.
2.
3.
4.
Purchase data
Awareness data
Competitive response
Source of sales
28
Test Markets
•
Simulated Test Market (STMs)
–
•
To gain insight into test marketing
Not actual test markets—rely instead on laboratory
Steps in the STM
1.
2.
3.
4.
5.
Intercept consumer at shopping malls
Screen
Expose to the new product concept or prototype
Give subject opportunity to buy new product
After time interval interview those who made a
purchase determine their assessment of it
6. Use the trial and repeat purchase for input into a
mathematical model
John Wiley & Son, Inc
29
Test Markets
•
Four Reasons for the Popularity of STMS
1.
2.
3.
4.
–
To gain insight into test marketing
Competitors less likely to know about test
STMs can be completed quickly
STMs are much cheaper than standard test
STMs can be very accurate
Other Types of Test Marketing
–
–
Rolling rollout
Lead country strategy
John Wiley & Son, Inc
30
SUMMARY
– What is an Experiment
– Demonstrating
Causation
– The Experimental
Setting
– Experimental Validity
– Experimental Notation
John Wiley & Son, Inc
– Extraneous Variables
– Experimental Design,
Treatment and Effects
– Limitations of
Experimental Research
– Selected Experimental
Design
– Test Markets
31
The End
Copyright © 2006 John Wiley & Son, Inc
John Wiley & Son, Inc
32