View
214
Download
0
Tags:
Embed Size (px)
Citation preview
2
Criteria for Establishing Causal Relationships
• Concomitant variation
• Temporal ordering of variables
• Control over other possible causal factors
3
Preliminary Definitions: Independent Variable
• Experimenter controlled variable
• Value can be manipulated by experimenters to whatever they wish it to be
4
Dependent Variable
• Value expected to be dependent on experimenter’s manipulation
• Criterion or standard by which results are judged
5
Test Units
Subjects or entities whose response to the experimental treatment are measured or observed
6
What is an Experiment?
• Research investigation under controlled conditions
• At least one independent variable is manipulated
• Its effect on at least one dependent variable is measured
• Purpose: To test hypothesis
8
Basic Experimental Design Issues
• Manipulation of independent variable
• Selection of dependent variable
• Assignment of subjects (or other test units) to treatments/conditions
• Control over extraneous variables
9
How Can Experimenters Control for Extraneous Variation?
• Eliminate Extraneous Variables
• Hold Conditions Constant
• Randomization
• Matching Subjects
• Presentation Order
• Blinding
15
Laboratory Experiment Field Experiment
High Internal Validity
Few ExtraneousVariables
High Control
Low Cost
Short Duration
Subjects Aware ofParticipation
High External Validity
Many ExtraneousVariables
Low Control
High Cost
Long Duration
Subjects Unaware ofParticipation
Demand Artifacts High Demand Artifacts Low
Few Participants Many Participants
Easier to Do Harder to Do
16
• Guinea pig effect
• Hawthorne effect
Demand Characteristics
Experimental procedures that intentionally hint to subjects something about experimenter’s hypothesis
17
Internal Validity
• Ability of experiment to determine if the experimental treatment was the sole cause of changes in a dependent variable
• Did the manipulation do what it was supposed to do?
18
Threats to Internal Validity
• History
• Maturation
• Testing
• Instrumentation
• Selection bias
• Mortality
19
Type of Extraneous Variable Example
History - Specific events in theenvironment between the Beforeand After measurement that are beyond the experimenter’s control
Maturation - Subjects changeduring the course of the experiment
Testing - The Before measure alertsor sensitizes subject to nature of experiment or second measure.
A major employercloses its plant intest market area
Subjects become tired
Questionnaireabout the traditionalrole of women triggers enhanced awareness of womenin an experiment.
20
Instrument - Changes ininstrument result in response bias
Selection - Sample selectionerror because of differentialselection comparison groups
Mortality - Sample attrition; some subjects withdraw from experiment
New questions aboutwomen are interpreteddifferently from earlierquestions.
Control group and experimental group isself-selected groupbased on preference forsoft drinks
Subjects in one groupof a hair dying study marry rich widows and move to Florida
Type of Extraneous Variable Example
35
Test Market
Field test of a new product or marketing mix element(s) of a current product using experimental or quasi-experimental designs
36
Uses and Objectives
• Evaluate proposed natural programs
• Estimate market shares and volumes
• Estimate cannibalization rate on existing product line
• Collect classification data on potential customers
• Analyze competitor reactions
37
Benefits and DisadvantagesBenefits
• Estimate sales potential under realistic conditions
• Find and correct weaknesses in marketing mix
Disadvantages
• Costs
• Tipping hand to competitors
38
What to Consider
• Tradeoff of costs and risk of product failure versus potential profits and probability of success
• Speed of competitor response
• Cost to produce product for test market
• Impact of failure on company reputation
39
Cost of Two-Market Test
Direct costs: $500,000
• POP materials, coupons, samples, syndicated research, media time/space, ad agency payments, initially high trade allowances
Indirect costs
• Management time, diversion from current products, negative internal/external impact of test failure, etc.
40
Steps in Test Market Study
1. Define objectives• Estimate market share, purchase frequency,
cannibalization rates• Discover who buys product and where
2. Select testing approach• Standard• Simulated (mall testing facility)• Controlled test market (e.g., BehaviorScan)
41
Steps in Test Market Study
3. Develop marketing plan
4. Select test markets• Should reflect regional differences if important,
but have similar demographics and little media spillover from/to other markets
• Avoid over-tested and idiosyncratic markets