Factorical ANOVA / Interaction Flashcards
Concept of interaction
The effect of core PV n the OV is different for different levels of moderator
How can interaction affect relationships
Difference in direction
Difference in size
Main effect of PV on x-axis
Check out direction of both lines
(If same direction there is main effect of that PV)
Main effect moderator PV
Check positioning of lines
(If one line always above the other, there is a main effect of MV)
Interaction effect
Check if lines have different slopes or are parallel
(If not parallel or have different slopes, there is an interaction effect)
Factorial ANOVA concept
Examining how much of the variance in our data can be explained by our predictor variables
(Full model and individual PV relationships)
Hypotheses
H0: miu 1 = miu 2 = … = miu i
H1: at least one miu differs
Requirements
1) one quantitative OV and 2 or more categorical PVs with 2 or more categories
2) independent mutually exclusive events
3) homogeneity of variances
4) normality
Which test
F test
Follow up tests
Simple contrast test
Helmet
Repeated contrast test
Interpreting moderation in ANOVA
Moderation: “When passing the PC lab (PV), attending the lecture (MV) leads to higher confidence (OV) than not attending”
No moderation: “Attending the lecture (MV) always leads to higher confidence (OV) what not attending”
Unstandardized beta interpretation
If we increase / decrease X by one unit we expect Y to increase / decrease by ____ units
Standardized beta interpretation
Example
“For every SD change in English test scores, the statistics exam score increases / decreases by ___ SDs”
Reporting mean difference
Mention mean difference value, confidence interval and p value