Deck 11: Moderation,mediation and More Regression

Question 1

Assume you wanted to investigate the length of time a team had played together (predictor variable) and the effect it had on the team's cohesion (outcome variable) with personality being the mediator variable. Which of the following best describes the direct relationship in this example?

A) Relationship between the length of time the team has been together and team cohesion
B) Relationship between the length of time the team has been together and team cohesion through personality
C) Relationship between the length of time the team has been together and personality
D) Relationship between personality and team cohesion

Answer 1

Relationship between the length of time the team has been together and team cohesion

Question 2

If you were researching the effect of hours training and attitude on the performance of a specific skill (e.g. goals scored) how many possible predictors would exist in attempting to determine the relationship between the predictor variable (hours of training), moderator variable (attitude) and the outcome variable (performance)?

A) 1
B) 2
C) 3
D) 4

Answer 2

3

Question 3

Which of the following statements best describes moderation?

A) The combined effect of two variables on another
B) The effect of a moderator variable on a predictor variable
C) The effect of an outcome variable on a moderator variable
D) The effect of a single variable on that of two others

Answer 3

The combined effect of two variables on another

Question 4

What is the statistical term for moderation?

A) Interaction effect
B) Between-group effect
C) Within-group effect
D) Repeated group effect

Question 5

If a researcher were interested in the determinants of muscular strength (outcome variable), which of the following could be considered as a predictor variables?

A) Concentration of testosterone
B) Sex of the participant
C) Number of years training
D) All of the options are appropriate

Question 6

If the following diagram illustrates mediation, what does the arrow marked as c′ represent?

A) The relationship between the predictor variable and the outcome variable following the introduction of the mediator variable
B) The relationship between the predictor variable and the outcome variable before the introduction of the mediator variable
C) The difference between the predictor variable and the outcome variable following the introduction of the mediator variable

Question 7

Based on the training example, how would the equation be expressed?

A) Performance = (Intercept +b₁training + b₂attitude + b₃interaction) + error
B) Performance = (Intercept - b₁training - b₂attitude - b₃interaction - b₄interaction) + error
C) Performance = (Intercept × b₁training × b₂attitude × b₃interaction × b₄interaction) + error
D) Performance = (Intercept + b₁training × b₂attitude × b₃interaction) - error

Question 8

What does the term 'centring' refer to?

A) The process of transforming a variable into deviations around a fixed point
B) The process of transforming a variable into whole numbers around a varied point
C) The process of transforming a variable into deviations around a variable point
D) The process of transforming a variable into whole numbers around a fixed point

Question 9

If you were investigating the moderator effect of attitude (i.e. positive or negative) and the relationship between hours training and performance, which of the following levels would be necessary to evaluate the interaction effects?

A) Low attitude scores
B) Average attitude scores
C) High attitude scores
D) All of the above

Question 10

How are the interaction scores calculated between the two variables (i.e. predictor and moderator)?

A) The scores of the two variables are multiplied together.
B) The scores of the two variables are divided by each other.
C) The scores of the two variables are added together.
D) The scores of the two variables are subtracted from each other.

Question 11

Which of the following definitions best describes mediation?

A) The relationship between a predictor variable and an outcome variable can be explained by a third variable.
B) The combined effect of two variables on another.
C) The ability to predict an outcome variable from two predictor variables.
D) The unexplained variance after the effect of two predictor variables has been examined

Question 12

Moderation is assessed using which statistical procedure?

A) Regression
B) Correlation
C) Discriminant analysis
D) Multilevel analysis

Question 13

Based on the training example, which of the following would be considered the second highest-order predictor?

A) Training × attitude
B) Training
C) Attitude
D) Performance

Question 14

Based on the training example, which of the following would be considered the highest-order predictor?

A) Training × attitude
B) Training
C) Attitude
D) Performance

Question 15

If c′ decreases in magnitude, what would this suggest?

A) Mediation has occurred and the relationship has become weaker between the predictor and outcome variable.
B) Mediation has not occurred but the relationship has become weaker between the predictor and outcome variable.
C) Mediation has occurred and the relationship has become stronger between the predictor and outcome variable.
D) Mediation has not occurred but the relationship has become stronger between the predictor and outcome variable.

Question 16

What would be the most likely term missing from this model?

A) Performance (e.g. number of goals scored)
B) Sex (M/F)
C) Age
D) Experience

Question 17

Assuming you were researching a positive attitude effect, how would you interpret the following result based on the relationship between training and performance? b = 0.511, 95% CI [0.191, 0.595], t = 3.98, p < .001

A) When the athlete's attitude is positive there is a positively significant relationship between training and sports performance.
B) When the athlete's attitude is positive there is a positively non-significant relationship between training and sports performance.
C) When the athlete's attitude is positive there is a negatively significant relationship between training and sports performance.
D) When the athlete's attitude is positive there is a negatively non-significant relationship between training and sports performance.

Question 18

In relation to the concept of moderation, which of the following best describes how an outcome variable is affected?

A) The interplay between a predictor and a moderator variable
B) The interplay between two or more dependent variables
C) The effect of a single predictor variable on a moderator variable
D) The effect of a single moderator variable on an outcome variable

Question 19

In using the example of training (predictor) and attitude (moderator) and the effect on sports performance (outcome variable), how would you determine whether moderation had occurred?

A) Hours training and attitude significantly predicted sports performance.
B) Hours training and attitude did not significantly predict sports performance.
C) Hours training significantly predicted sports performance.
D) Attitude significantly predicted sports performance.

Question 20

If mediation has taken place between the predictor and the outcome variable, what is likely to happen to the relationship?

A) The strength of the relationship will have decreased.
B) The strength of the relationship will have increased.
C) The strength of the relationship may increase or decrease.
D) The strength of the relationship should remain constant.

Question 21

Which of the following best describes the indirect relationship in this example?

A) Relationship between the length of time the team has been together and team cohesion
B) Relationship between the length of time the team has been together and team cohesion through personality
C) Relationship between the length of time the team has been together and personality
D) Relationship between personality and team cohesion

Question 22

Suppose a kappa-squared test was used to measure the size of the indirect effect in the cohesion example. What would a value of .1 suggest?

A) Small effect
B) Medium effect
C) Large effect
D) The largest possible effect

Question 23

A regression coefficient for an indirect effect is usually expressed in an unstandardized format. Why might it be useful to convert the value to a standardized regression format?

A) To enable a comparison across studies.
B) It improves the accuracy of the calculation.
C) The confidence intervals are reduced.
D) All of the options are applicable.