Week 7 Flashcards
enhancing
moderating IV strengthens association between IV and DV
buffering
moderating IV reduces association between IV and DV
antagonistic
moderating IV flips direction of association between IV and DV
interacting variable with itself is a sneaky way that we can
introduce some nonlinearity into linear regression models
ANOVA assumptions
Continuous DV
one IV is continuous (if both are categorical, do factorial ANOVA)
independence of observations
no multicollinearity
not significant outliers
homogeneity of variance
residual variance in the outcome that remains after predicting Y from X is the same for different values of M
homogeneity of variance if M is dichotomous
category with smaller variance will have weaker slop
larger variance stronger slope, transform X or M