Suzanne van Heyningen Experimental Research
Summary Experimental Research
Last exam tips
- Articles and book are additional material
- If a number can be higher than 1 you need to write a 0, e.g. p = .05 whereas F = 0.30
- If you have a significant effect, always report M’s and SD’s
- If you talk about means in your hypothesis, you also need to report it
1
,Suzanne van Heyningen Experimental Research
Lecture one
Conversion optimization
A structured and systematic Taking the traffic there is
Informed by data
approach to improving the and making the most of it
insights & psychology
performance of a website
Types of Marketing Research
Survey: Experiment:
a) Cross- sectional a) Field
b) Longitudinal b) Lab
c) Panel
Meaning • Gather information with • Factor is isolated to test hypothesis
questionnaire • 1 or more IV’s are manipulated and
change DV while controlling for
extraneous variable
Knowledge focus • Descriptive • Experimental
• Hypothesis testing • Hypothesis testing
Validity focus • External validity • Internal validity
Breath of finding • Medium • Narrow
Why experimenting?
1. Describe
2. Predict
3. Explain behavior of market-parties/ employees/ buyers
❖ Break up phenomena in variables and relations between those variables
Behavioral Research
1. Descriptive research
- Thoughts, feelings, ideas, behaviors
2. Correlational research
- Identifying relationships between different observed variables: measuring thoughts, feelings,
behavior
- It only shows a relationship but it cannot tell whether one variable actually causes the other
- Spurious correlation – third explaining variable
2
,Suzanne van Heyningen Experimental Research
From direction to prediction
1. Description
- Careful observation
2. Correlation
- relationship between observed variables
3. Experimental
- When the quest for causality is the objective, then an experiment is the method of choice
- Example : What impact does assortment size have on store choice?
Causality
Relationship between Asymmetrical A (cause) →
2 variables direction B (effect)
Change in A is No alternative explanation for
accompanies by change in B than change in A
change in B (control others)
Experimental Research
1. It is the only type of research that (potentially) can demonstrate that change in one variable causes a
predictable change in another variable.
2. Most difficult: making sure that a change in Y was not caused by something else than X.
- Test specific hypotheses about the relationship between a cause and an effect via controlled
The effect of an IV on a DV
- Manipulating the independent variable
- Measure effects on dependent variable
- Control other influences (high internal but low external validity)
Importance of randomization
Use different people with Potential confounds Change in Y can
different characteristics are under control be attributed to X
Experimental research : Main steps
1. Theoretical Framework
- Problem Identification
- Hypothesis Formulation
2. Experimental Design
- Manipulation of IVs
- Measurement of DV
- Control for confounds
3. Data analyses
- Get familiar with your data
- Checks
- Conduct main analyses
- Conduct follow up analyses
3
, Suzanne van Heyningen Experimental Research
Problem Identification
Sources of research ideas
• Real life experiences
• Previous research and theory
- Conflicting findings
- Boundary conditions
- Find explanation for observed effects
- Propositions of a theory that are not tested yet
- Applying a theory to a consumer setting
Define your topic by identifying all relevant factors
- DV : What do you want to predict/explain/measure?
- IV : What do you think is the cause/manipulate?
- Moderator : What do you think can change the effect?
- Mediator : What do you think drives the effect?
Define problem statement
- Always draw a conceptual model
Good research problems should both
- Have real life relevance (implication for managers and consumers)
- Contribute to current knowledge (theoretical contribution)
4
Summary Experimental Research
Last exam tips
- Articles and book are additional material
- If a number can be higher than 1 you need to write a 0, e.g. p = .05 whereas F = 0.30
- If you have a significant effect, always report M’s and SD’s
- If you talk about means in your hypothesis, you also need to report it
1
,Suzanne van Heyningen Experimental Research
Lecture one
Conversion optimization
A structured and systematic Taking the traffic there is
Informed by data
approach to improving the and making the most of it
insights & psychology
performance of a website
Types of Marketing Research
Survey: Experiment:
a) Cross- sectional a) Field
b) Longitudinal b) Lab
c) Panel
Meaning • Gather information with • Factor is isolated to test hypothesis
questionnaire • 1 or more IV’s are manipulated and
change DV while controlling for
extraneous variable
Knowledge focus • Descriptive • Experimental
• Hypothesis testing • Hypothesis testing
Validity focus • External validity • Internal validity
Breath of finding • Medium • Narrow
Why experimenting?
1. Describe
2. Predict
3. Explain behavior of market-parties/ employees/ buyers
❖ Break up phenomena in variables and relations between those variables
Behavioral Research
1. Descriptive research
- Thoughts, feelings, ideas, behaviors
2. Correlational research
- Identifying relationships between different observed variables: measuring thoughts, feelings,
behavior
- It only shows a relationship but it cannot tell whether one variable actually causes the other
- Spurious correlation – third explaining variable
2
,Suzanne van Heyningen Experimental Research
From direction to prediction
1. Description
- Careful observation
2. Correlation
- relationship between observed variables
3. Experimental
- When the quest for causality is the objective, then an experiment is the method of choice
- Example : What impact does assortment size have on store choice?
Causality
Relationship between Asymmetrical A (cause) →
2 variables direction B (effect)
Change in A is No alternative explanation for
accompanies by change in B than change in A
change in B (control others)
Experimental Research
1. It is the only type of research that (potentially) can demonstrate that change in one variable causes a
predictable change in another variable.
2. Most difficult: making sure that a change in Y was not caused by something else than X.
- Test specific hypotheses about the relationship between a cause and an effect via controlled
The effect of an IV on a DV
- Manipulating the independent variable
- Measure effects on dependent variable
- Control other influences (high internal but low external validity)
Importance of randomization
Use different people with Potential confounds Change in Y can
different characteristics are under control be attributed to X
Experimental research : Main steps
1. Theoretical Framework
- Problem Identification
- Hypothesis Formulation
2. Experimental Design
- Manipulation of IVs
- Measurement of DV
- Control for confounds
3. Data analyses
- Get familiar with your data
- Checks
- Conduct main analyses
- Conduct follow up analyses
3
, Suzanne van Heyningen Experimental Research
Problem Identification
Sources of research ideas
• Real life experiences
• Previous research and theory
- Conflicting findings
- Boundary conditions
- Find explanation for observed effects
- Propositions of a theory that are not tested yet
- Applying a theory to a consumer setting
Define your topic by identifying all relevant factors
- DV : What do you want to predict/explain/measure?
- IV : What do you think is the cause/manipulate?
- Moderator : What do you think can change the effect?
- Mediator : What do you think drives the effect?
Define problem statement
- Always draw a conceptual model
Good research problems should both
- Have real life relevance (implication for managers and consumers)
- Contribute to current knowledge (theoretical contribution)
4
