Final-lecture 15

Question 1

Define moderation

Answer 1

• combined effect of two variables on another variable

- interaction effect

Question 2

How does a moderator variable effect X and Y?

Answer 2

-it affects the relationship between the IV (predictor) and the DV (outcome)

Question 3

How can you tell if two variables are truly interacting?

Answer 3

-if the effect of one of the variables differs depending on the level of the other variable

Question 4

Give an example of moderation?

Answer 4

• the effect of sex on marriage and life satisfaction

- men who’re married are more satisfied

Question 5

Why use interactions?

Answer 5

• the social world is complex
• a combination of the interaction matters
• tells us more accurate depictions of real world

Question 6

Do interactions include mediation?

Answer 6

NO just moderation

Question 7

What is the interaction of two variables equal to?

Answer 7

-the scores of the two variables multiplied together

Question 8

What determines whether moderation has occurred?

Answer 8

• a significant interaction effect

- the predictor and moderator variables do not need to be significant

Question 9

What must we include for the interaction model and term to be valid?

Answer 9

-the predictor and the moderation in the model

Question 10

How is moderation observed in a regression?

Answer 10

-measuring the outcome when the model includes a predictor, a moderator and the interaction of these two variables

Question 11

What is the first step of an interaction model?

Answer 11

-ensuring the IV and DV are significant

Question 12

If we are looking at religious attendance and depression and how race affects this what are we theoretically saying?

Answer 12

-race will affect the relationship between attendance and depression

Question 13

What is analogous to testing for moderation?

Answer 13

-t-test of difference in group means

Question 14

What is the null hypothesis in a t-test?

Answer 14

-there is no difference between the two group means in the population

Question 15

Using the example of race, depression and religious attendance what would our null hypothesis be?

Answer 15

-the slope for relationship between attendance and depression is the same for the two race groups

Question 16

Testing group differences in b is like what?

Answer 16

-a t-test of means

Question 17

What is the b for an interaction using our example of race?

Answer 17

-the difference in the coefficient for attendance between white and black respondents

Question 18

If the interaction term is not significant, what do we do?

Answer 18

-not include it in our model

Question 19

For multicollinearity, what are we concerned with?

Answer 19

-high correlations between predictor variables not the dependent variable