Chapter 11 Flashcards
Define interaction of two independent variables
Whether the effect of the original independent variable depends on another independent variable
Interaction = a difference in differences
Define crossover interaction
An interaction where the lines on the graph cross each other
Define spreading interaction
Lines are not parallel and do not cross over
Define a factorial design
Experiment design
One in which there are two or more independent variables.
Tests all possible combinations of conditions
If a study has two IV and each IV has two levels there will be four conditions
Define nested factorial designs
One independent variable is primary, and the other independent variable is nested under it
(Ex: researchers might study two school districts and then study three schools within each district)
Define participant variable
A variable whose levels are selected, not manipulated
Ex: age, sex, race
Define moderator
(Interactions show moderators)
An independent variable that changes the relationship between another independent and dependent variable
(Results in an interaction)
Define a main effect
The overall effect of one independent variable on the dependent variable, averaging over the levels of the other independent variable
Define marginal means
The means for each level of an independent variable, averaging over levels of the other independent variable
If lines are not parallel in a line graph then there is probably…
An interaction
Can’t confirm without significance test
In a bar graph if the bars are the same height then there probably is
No interaction
Can’t confirm without test
What’s the best way to describe an interaction
Start with one level of the first independent variable and explain what is happening with the second independent variable at that level, and then moved to the next level of the first independent variable and do the same thing
Ex: when people poured from the small bottle, they poured more detergent when it was on sale than when it was regular priced. When people poured from the large bottle, they poured the same amount of detergent when it was on sale as one it was regular price
What’s more important the main effect or an interaction
Interaction
If lines are not parallel in a line graph then there is probably…
An interaction
Can’t confirm without significance test
In a bar graph if the bars are the same height then there probably is
No interaction
Can’t confirm without test
What’s the best way to describe an interaction
Start with one level of the first independent variable and explain what is happening with the second independent variable at that level, and then moved to the next level of the first independent variable and do the same thing
Ex: when people poured from the small bottle, they poured more detergent when it was on sale than when it was regular priced. When people poured from the large bottle, they poured the same amount of detergent when it was on sale as one it was regular price
What’s more important the main effect or an interaction
Interaction
Define within groups
The same people participate at each level
One group participates in all 4 combinations
Define between groups (independent-groups)
Different people participate at each level
Define mixed factorial design
One independent variable is manipulated as independent groups (between) and the other is manipulated as within groups
Why do we use factorial design’s
It has better external validity
Typically, our behavior depends on more than one variable by testing multiple factors we better approximate the “normal” conditions
(Ex: plant growing w both water and sunlight)
But we are limited with how many factors we can include (# of participants, complexity of data analysis)
Why do we use factorial design (dependencies)
Many Times the answers to our research questions is it depends
(Ex: best therapy depends on the patient)
Factorial designs allows us to assess these dependencies
Why use factorial designs? (Interaction)
If the effect of one variable depends on another the variables are said to “interact”
Factorial designs allows us to test for interactions
(Ex: does the effect of light depend on how much water the plant has)
How do you describe the findings and there’s not an interaction
Ex: there’s not an interaction between them out of coffee drunk and I’m out of beer drunk. The fact of how much beer was drunk was the same no matter how much coffee was drunk
How do you calculate an interaction
Calculate the size of the effect of one variable (by subtracting one value from the other) for each level of the other variable
If they’re the same then there is no interaction
(Find differences)
Describing a main effect
Ex: There was a main effect of coffee.
On average, participants who drank two cups of coffee slept 2 hours less than participants who drank no cups of coffee
How do you calculate main effect
Add and divide sections it asks about and compare sum (find avg)
If sum is the same there is no main effect
