Association
- if there is a causal relationship between IV and DV, there must be a strong correlation between them
Temporal Priority
- IV must be an antecedent of the DV
- i.e. IV --> DV
Controlling for Common-causal (or lurking) Variable
- the influence of common-causal variables that may have produced spurious relationships between IV and DV should be ruled out
Two requirements for a business experiment
- a control group
- feedback mechanism for you to be able to observe the effects
Advantages and Disadvantage of Repeated Measures
- advantages: increase statistical power, economize participants
- disadvantages: the first measure will influence the second measure (carryover effect), participants might become fatigued, participants performance might improve over time through practice
Factorial Design
- two or more IV's, each with two or more levels
- all possible combinations of levels A and B
Tactical Segmentation
- based on the response to tactical moves that you do for your company
- if two controllable marketing decision variables interact, implication is that you must coordinate
Mediator Varibales
- account for the relationship between a predictor (IV) and the dependent variable (DV)
- theoretically, X causes the mediator variable, which causes Y
- if the effect of the IV on the DV decreases dramatically when another lurking variable is introduced,
Moderation
- moderator variables change the impact of one variable on another
- conceptually similar to the 'interaction' term
- if the effect of the IV on the DV is different at different partitioned levels of another lurking (hidden) variable, a moderator is prese
Presence of a main effect
- if independently the A and B have an impact on the DV
Presence of interaction effect
- if A and B have a joint impact on the DV
Surveys
- used for descriptive research (not causal)
- provide a snapshot at a point in time
Double Barrel Questions
- questions that asks two questions in one statement
- ex. do you think Coke is a tasty and refreshening soft drink? This is double barrel because some people might think Coke is tasty but not refreshening or the other way around
Funnel Order
- ask questions in order of general to specific
- warms the respondent up
Inverted Funnel Order
- ask questions in order of specific to general for complex topics
Semantic Differential
- typically 7 point scale (on a scale of 1-7)
- typically bipolar adjectives
Likert Scale
- agree - disagree
- usually a 5 or 7 point scale
- can be used item by item for attributes
- can be used for comparisons across two brands
Constant Sum
- often used to measure importance
- generally do not want to have more than 5-7 features
- ex. how important is it that your on-the-go breath freshener provides you the following set of benefits? Please allocate 100 points amongst the benefits shown
Purchase Intent
- statement of likelihood of making a future purchase