Deck 11: Moderation, Mediation and More Regression

Question 1

A busy social life has been found to increase happiness in participants who are experiencing low levels of stress, but decrease happiness in participants who are experiencing high levels of stress. What is this an example of?

A)Semi-partial correlation
B)Moderation
C)Mediation
D)Neither moderation nor mediation

Answer 1

B

Question 2

Mediation has occurred when:

A)The relationship between two variables changes as a function of a third variable.
B)The strength of the relationship between a predictor variable and an outcome variable is reduced by including another variable in the model.
C)The strength of the relationship between a predictor variable and an outcome variable is increased by including another variable as a predictor.
D)The relationship between two variables decreases as a function of a third variable.

Answer 2

B

Question 3

Imagine we wanted to investigate whether a person's profession can predict scores on a self-report psychopathy scale. We collected data from people in eight professions and a group of unemployed people. The eight professions were: bank traders, insurance brokers, health care professionals, business executives, volunteer workers, full-time mums, teachers, construction workers. The outcome was psychopathy score. How could we analyse these data?

A)Regression only
B)ANOVA or chi-square
C)ANOVA only
D)ANOVA or regression

Answer 3

D

Question 4

What is the Sobel test used for?

A)To assess the significance of the indirect effect in mediation analysis
B)To assess the significance of the direct effect in mediation analysis
C)To assess the significance of the interaction effect in moderation analysis.
D)To assess the significance of the direct effect in moderation analysis

Question 5

Imagine we found a strong positive correlation between worry and sleep disturbances and we hypothesized that drinking caffeine before going to bed would exacerbate this relationship. What type of analysis could we conduct to test this hypothesis?

A)Moderation
B)Mediation
C)ANCOVA
D)Two-way ANOVA

Question 6

Imagine we wanted to look at predictors of popularity of famous sports people. We might hypothesize that the type of sport played would predict how popular the sports person was. Imagine we recruited 1000 participants and asked them to rate out of 10 (10 = They are the best sports person in the world, 0 = They are the worst sports person in the world) how much they liked 100 famous sports people from 10 different types of sports (10 sports people from each category). Could we analyse these data using regression, and if so how?

A)Yes.To do this we would create 9 dummy variables for the 'type of sport played' variable.
B)Yes.To do this we would need to create one coding variable for the 'type of sport played' predictor variable.
C)No.We could not conduct a regression because our categorical predictor is made up of more than two categories.
D)No.We can only analyse this scenario using ANOVA.

Question 7

Imagine we conducted a mediation analysis and obtained a kappa-squared value of .8 for the indirect effect. What does this tell us?

A)That the indirect effect is medium relative to the maximum possible value.
B)None of these
C)That the indirect effect is close to being as large as it could possibly be given the research design.
D)That the indirect effect is small relative to the maximum possible value.

Question 8

Which of the following is not one of Baron and Kenny's four conditions of mediation?

A)The predictor variable must predict the outcome variable when the mediator is entered into the analysis.
B)The predictor variable must significantly predict the outcome variable.
C)The predictor variable must significantly predict the mediator variable.
D)The mediator must significantly predict the outcome variable.

Question 9

Which of the following sentences best describes mediation?

A)Mediation refers to a situation in which the relationship between a predictor variable and an outcome can be explained by their relationship to a third variable.
B)Mediation refers to the combined effect of two variables on an outcome.
C)Mediation refers to a situation in which the relationship between a predictor variable and an outcome is significant only when the mediator is included in the model.
D)Mediation refers to a situation in which the relationship between a predictor variable and an outcome is unaffected by their relationship to a third variable.

Question 10

Imagine it was found that worry explains the relationship between depression and sleep problems (i.e., depressed people tend to worry more than non-depressed people, leading them to experience more sleep problems). Which of the following is the indirect effect?

A)The effect of sleep problems on depression through worry
B)The relationship between worry and sleep problems, controlling for depression
C)The effect of depression on sleep problems through worry
D)The relationship between depression and sleep problems, controlling for worry

Question 11

Imagine we wanted to investigate whether a person's profession can predict scores on a self-report psychopathy scale. We collected data from people in eight different professions and a group of unemployed people. The eight professions were: bank traders, insurance brokers, health care professionals, business executives, volunteer workers, full-time mums, teachers, construction workers. The outcome was psychopathy score. If we wanted to analyse these data using regression, how many dummy variables would we need?

A)7
B)It is not possible to analyse these data with regression.
C)8
D)9

Question 12

Imagine we found that drinking caffeine before going to bed significantly moderates the relationship between worry and sleep disturbances. What does this tell us?

A)That the relationship between worry and sleep problems changes as a function of caffeine consumption.
B)None of these.
C)That drinking caffeine before going to bed increases worry which in turn increases sleep disturbances.
D)That people who worry are more likely to drink caffeine before going to bed, leading to sleep disturbances.

Question 13

A simple slopes analysis:

A)Quantifies the rate of decrease, or 'slope', of the direct effect when a mediator is present.
B)Looks at the relationship between a predictor and outcome variable at low, mean and high levels of a moderator.
C)Is the slope of the regression line between a predictor and outcome variable.
D)Compares regression slopes within different participants.

Question 14

Which of the following is an example of perfect mediation?

A)The strength of the relationship between the predictor and the outcome is reduced by exactly half when the mediator is included in the model.
B)The relationship between the predictor and the outcome remains the same when the mediator is included in the model.
C)The interaction of the predictor and the mediator significantly predicts the outcome, but the variables themselves do not.
D)The relationship between the predictor and the outcome is completely wiped out when the mediator is included in the model.

Question 15

Which of the following sentences about grand mean centring in moderation analysis is not true?

A)Centring refers to the process of transforming a variable into deviations around a fixed point.
B)Grand mean centring for a given variable is achieved by taking each score and subtracting from it the mean of all scores (for that variable).
C)Centring is particularly important when your model contains an interaction term.
D)Centring the predictors will directly affect the b for the highest-order predictor, but will have no effect on the bs for the lowest-order predictors.

Question 16

How do we test for moderation?

A)Look at the effect of the moderator variable in a regression in which the outcome is predicted from a predictor and the proposed moderator.
B)Look at the indirect effect of a predictor on an outcome variable via the proposed moderator.
C)Look at the reduction in the direct effect of a predictor on an outcome variable when the proposed moderator is also included.
D)Look at the interaction term in a regression in which the outcome variable is predicted from a predictor, the proposed moderator and their interaction.

Question 17

An experiment was done to look at whether different relaxation techniques could predict sleep quality better than nothing. A sample of 400 participants were randomly allocated to one of four groups: massage, hot bath, reading or nothing. For one month each participant received the same relaxation technique for 30 minutes before going to bed each night. A special device was attached to the participant's wrist that recorded their quality of sleep, providing them with a score out of 100. The outcome was the average quality of sleep score over the course of the month.

- Which test could we use to analyse these data?

A)Regression or ANOVA
B)Regression only
C)ANOVA only
D)Chi-square

Question 18

An experiment was done to look at whether different relaxation techniques could predict sleep quality better than nothing. A sample of 400 participants were randomly allocated to one of four groups: massage, hot bath, reading or nothing. For one month each participant received the same relaxation technique for 30 minutes before going to bed each night. A special device was attached to the participant's wrist that recorded their quality of sleep, providing them with a score out of 100. The outcome was the average quality of sleep score over the course of the month.

-Which of the following tables displays the most appropriate coding scheme for analysing these data?

A)
B)
C)
D)

Question 19

Imagine we found a strong positive correlation between depression and sleep problems. We might hypothesize that this relationship is explained by worry (i.e., depressed people tend to worry more than non-depressed people, leading them to experience more sleep problems). What type of analysis could we conduct to test this hypothesis?

A)Moderation
B)ANCOVA
C)Loglinear analysis
D)Mediation

Question 20

Why is it common in moderation analysis to transform the predictors using grand mean centring?

A)Because it makes the bs for lower-order effects interpretable.
B)Because it makes the b for the higher-order interaction effect interpretable.
C)To correct for heteroscedasticity in the data.
D)It makes the data normally distributed.

Question 21

A researcher had a categorical variable that they wanted to include as a predictor in a regression equation. The researcher was trying to predict the success of a back pain intervention, and the categorical variable was the duration of the back pain prior to treatment, with 4 categories: less than 6 months, 6-12 months, 1-2 years, more than 2 years. They needed to code these variables into dummy variables for the regression using less than 6 months as their control category. Which of the following represents the correct coding scheme?

A) $$\begin{array} { | l | l | r | r | r | } \hline\quad\quad\quad\quad & \quad\quad\begin{array} { c } \\\\\\ \text { Duration of Pain } \\\end{array} & \begin{array} { c } \text { Dummy 1 } \\\text { (Under 6 } \\\text { Months vs } \\\text { 6-12 Months) }\end{array} & \begin{array} { c } \text { Dummy 2 } \\\text { (Under 6 } \\\text { Months vs } \\1 - 2 \text { Years) }\end{array} & \begin{array} { c } \text { Dummy 3 } \\\text { (Under 6 } \\\text { Months vs } \\\text { Over 2 Years) }\end{array} \\\hline & \text { Under 6 Months } & 1 & 1 & 1 \\& \text { 6-12 Months } &1 & 0 & 0\\& 1 - 2 \text { Years }& 0 & 1 & 0 \\&\text { More Than 2 Years }& 0 & 0 & 1\\ \hline\end{array}$$
B) $$\begin{array} { | l | l | r | r | r | } \hline\quad\quad\quad\quad & \quad\quad\begin{array} { c } \\\\\\ \text { Duration of Pain } \\\end{array} & \begin{array} { c } \text { Dummy 1 } \\\text { (Under 6 } \\\text { Months vs } \\\text { 6-12 Months) }\end{array} & \begin{array} { c } \text { Dummy 2 } \\\text { (Under 6 } \\\text { Months vs } \\1 - 2 \text { Years) }\end{array} & \begin{array} { c } \text { Dummy 3 } \\\text { (Under 6 } \\\text { Months vs } \\\text { Over 2 Years) }\end{array} \\\hline & \text { Under 6 Months } & 0 & 0 & 0 \\& \text { 6-12 Months } &1 & 0 & 0\\& 1 - 2 \text { Years }& 0 & 1 & 0 \\&\text { More Than 2 Years }& 0 & 0 & 1\\ \hline\end{array}$$
C) $$\begin{array} { | l | l | r | r | r | } \hline\quad\quad\quad\quad & \quad\quad\begin{array} { c } \\\\\\ \text { Duration of Pain } \\\end{array} & \begin{array} { c } \text { Dummy 1 } \\\text { (Under 6 } \\\text { Months vs } \\\text { 6-12 Months) }\end{array} & \begin{array} { c } \text { Dummy 2 } \\\text { (Under 6 } \\\text { Months vs } \\1 - 2 \text { Years) }\end{array} & \begin{array} { c } \text { Dummy 3 } \\\text { (Under 6 } \\\text { Months vs } \\\text { Over 2 Years) }\end{array} \\\hline & \text { Under 6 Months } & 0 & 0 & 0 \\& \text { 6-12 Months } &0 &1 & 1\\& 1 - 2 \text { Years }& 1 & 0 & 1 \\&\text { More Than 2 Years }& 1 &1&0\\ \hline\end{array}$$

D) $$\begin{array} { | l | l | r | r | r | } \hline\quad\quad\quad\quad & \quad\quad\begin{array} { c } \\\\\\ \text { Duration of Pain } \\\end{array} & \begin{array} { c } \text { Dummy 1 } \\\text { (Under 6 } \\\text { Months vs } \\ \text { 6-12 Months) }\end{array} & \begin{array} { c } \text { Dummy 2 } \\\text { (Under 6 } \\\text { Months vs } \\1 - 2 \text { Years) }\end{array} & \begin{array} { c } \text { Dummy 3 } \\\text { (Under 6 } \\\text { Months vs } \\\text { Over 2 Years) }\end{array} \\\hline & \text { Under 6 Months } & 0 & 0 & 0 \\& \text { 6-12 Months } &0 &1 & 1\\& 1 - 2 \text { Years }& 1 & 0 & 1 \\&\text { More Than 2 Years }& 1 &1&0\\ \hline\end{array}$$