Ch12 Flashcards
interaction effects
Interaction effects (interaction): the effect of an independent variable depends on the level of another
type of interactions
- crossover
-spreading
Crossover interaction
Crossover interaction: “it depends”- (ex: ice cream cold, pancakes hot) looks like x on graph
spreading interaction
Spreading: “only when”- (ex: dog sits when told, only when there’s a treat) looks like <
Factorial design
Factorial design: studies two or more IVs (factors)
Cells
unique conditions representing combinations of IV’s
participant variable
Participant variable: in factorial design- a variable that’s levels are selected (measured but not manipulated) but act in place of a second IV (ex: age can’t be manipulated)
moderators
Moderators (in factorial design): IV that changes relationship between another IV and the DV (results in an interaction)
moderators
Moderators (in factorial design): IV that changes relationship between another IV and the DV (results in an interaction)
main effect
Main effect: the effect of one independent variable on the DV, if you avg over/ignoring the levels of the other IV
marginal means
Marginal means: means for each level of an IV if you avg over levels of the other IV
computing interactions
- Start with one level of IV1, compute the difference btwn the levels of IV2 (using subtraction), then compute difference btwn levels of IV2 for the other level of IV1
- Then compare the numbers for each difference. If they are different enough it is statistically significant, there is an interaction
Independent-groups factorial designs (between-subjects…)
both IV’s are studied as independent groups, so each cell has it’s own group of people
Within-groups factorial designs (repeated measures…):
Within-groups factorial designs (repeated measures…): all participants are part of all of the conditions/cells
mixed factorial design:
Mixed factorial design: one IV is manipulated as independent groups and the other is within-groups
(Intermediate # of participants)
factorial design notation
the number tells how many levels of an IV, and the # of #’s tell how many IV’s
2 x 2 x 2 factorial or three-way design
- two levels of each IV, 3 IVs
- 8 cells ((2x2)x2)
- 3 main effects to test, 3 two way interactions (avg over third v), 1 three way interaction
how to depict three-way factorial design
Depict by constructing table twice, once for each level of the third IV, two side by side graphs
what does it mean if a three way interaction is significant
- means the two-way interaction between two of the IVs depends on the level of the third IV
how to tell if 3-way interaction
How to know if there is a three way interaction:
- If there is a two-way interaction for one level of the third v but not the other, OR
- If there are two different patterns of two-way interactions for the levels of the third v
where to look in journal articles to see if design is factorial
Method section: find info about study design and variables
- _ x_ indicates factorial design, along with numbers of IVs and levels
- Also can see if within-groups or independent-groups, mixed factorial design
Results section: shows whether main effects and interactions are significant
- Significance, p value ( p < 0.05) or *
- MANOVA or ANOVA or F indicate factorial interactions
in media, look to see if factorial
- look for “It depends” or “only when”
- Look for participant variables (age, gender, ethnicity, etc.)
It might moderate another variable, and when there’s a moderator then there’s an interaction, a factorial design
