Topic 5: Interaction & Mediation

Question 1

interaction effect

Answer 1

the extent to which the effect of one factor depends on the level of the other factor

Question 2

when is an intreaction present?

Answer 2

when the effect of one factor on the DV changes at different levels of the other factor

Question 3

interpreting cell mean plots

Answer 3

• if the slopes are the same, there is no interaction
• if the slopes are different (& the lines eventually intersect), there is an interaction

Question 4

interactions in linear regression

Answer 4

a change in one predictor’s relationship with the DV when another predictor changes

Question 5

three types of interactions in linear regression

Answer 5

1. interaction b/n two continous predictors
2. interaction b/n nominal & continuous predictors
3. interaction b/n nominal predictors (two-way ANOVA)

Question 6

interaction between two continuous predictors formula

Answer 6

Ŷ = a + B1X1 + B2X2 + B3X1X2

Question 7

a (interaction b/n two continuous predictors formula)

Answer 7

average in y when x1 = x2 = 0

Question 8

b1 (interaction b/n two continuous predictors formula)

Answer 8

effect of x1 when x2 = 0

Question 9

b2 (interaction b/n two continuous predictors formula)

Answer 9

effect of x2 when x1 = 0

Question 10

b3 (interaction b/n two continuous predictors formula)

Answer 10

change in the effect of x1 on average as x2 increases by 1 unit

Question 11

H0 (interaction b/n two continuous predictors formula)

Answer 11

B3 = 0 (no interaction b/n x1 & x2)

Question 12

H1 (interaction b/n two continuous predictors formula)

Answer 12

B3 ≠ 0 (x1 & x2 interact)

Question 13

interaction b/n binary & continuous predictors formula

Answer 13

we use the same regression equation as for two continuous predictors

Question 14

interaction b/n multicategorical & continuous predictors formula

Answer 14

Ŷ = a + B1X1 + B2D1 + B3D2 + B4X1D1 + B5X1D2

Question 15

a (interaction b/n multicategorical & continuous predictors)

Answer 15

average for y for group 3 (baseline) when x1 = 0

Question 16

b1 (interaction b/n multicategorical & continuous predictors)

Answer 16

effect of x1 on y for group 3

Question 17

b2 (interaction b/n multicategorical & continuous predictors)

Answer 17

effect of D1 on y when x1 = 0

Question 18

b3 (interaction b/n multicategorical & continuous predictors)

Answer 18

effect of D2 on y when x1 = 0

Question 19

b4 (interaction b/n multicategorical & continuous predictors)

Answer 19

differences in slopes for x1 b/n groups 1 & 3

Question 20

b5 (interaction b/n multicategorical & continuous predictors)

Answer 20

differences in slopes for x1 b/n groups 2 & 3

Question 21

hierarchical principle

Answer 21

• if we include an interaction term in a model, we should also include the main effects, even if they’re not statistically significant
• ex. x1x2 is normally correlated with x1 & x2

Question 22

mean-centring

Answer 22

• mean-centring x1 & x2 gives us a meaningful way of interpreting their regression coefficients
• the means of centred x1 & x2 = 0
• B1 indicates the effect of x1 on y among those average on x2 & vice versa

Question 23

3 types of effects in mediation

Answer 23

direct, indirect, and total

Question 23

mediation analysis in linear regression

Answer 23

used to quantify pathways of influence or the process by which an IV can influence a DV

Question 24

direct effect

Answer 24

the influence of one variable on another that isn’t mediated by any other vairable

Question 25

indirect effect

Answer 25

the influence of a variable mediated by at least one intervening variable

Question 26

total effect

Answer 26

direct effect + indirecteffect

Question 27

simple mediation model

Answer 27

• total effect of x1 on y
• direct & indirect effects of x1

Question 28

simple mediation model with covariates

Answer 28

all of the interpretations of total, direct, and indirect effects remain the same, but with the addition of holding x3 & x4 constant

Question 29

inference about indirect effect

Answer 29

testing the statistical significance of unstandardized indirect effects with a single mediator

Question 30

two tests of unstandardized indirect effects with a single mediator

Answer 30

bootstrap CI, Monte Carlo Ci