lecture 26 (stats)

Question 1

moderation?

Answer 1

• an interaction effect between the predictor variable and the moderator (when one predictor variable influences the effects of the other predictor variable)
• a moderator is a variable that affects the direction and strength of the relationship between the independent variable and the dependent variable
• if the moderator present is continuous the slope of X and Y will change as the moderator changes (3D model)

Question 2

linear regression formula?

Answer 2

Out = b0 + b1Pred + b2Mod + b3PredMod + error

• check notebook for model

Question 3

check multicollinearity?

Answer 3

• can be done through correlation matrix -> check correlation between two predictor variables
• still prefereble to use Tolerance and VIF in Jasp

Question 4

interpretation of the model?

Answer 4

• the better the correlation between expectated and observed values the better our model performed
• r^2: propotion of explained variance

Question 5

how to do moderation analysis in Jasp?

Answer 5

• linear regression tab
• put dependent variables and covariates at the right place
• go to model -> select both independent variables -> add them to model 1 -> you get the interaction effect on the tables
• also possible to do it in process (beta) tab but use process beta in exam

Question 6

hierarchical regression in jasp?

Answer 6

• use it if you want to isolate the interaction effect
• add predictor variables without interaction effect to null model -> go to statistics -> model summary -> R squared change (how much did R^2 increased by looking at the interaction effect

Question 7

how to visualize interactioneffect on continous variables in Jasp?

Answer 7

descriptives -> flexplot -> put dependent variable and independent variable in respective place -> options -> fitted line -> regression