PSYC232 Test 3, Week 9

Question 1

What kind of research questions does moderation help answer?

Answer 1

“It depends” questions—specifically when, who, or where something happens.

Question 2

What is moderation in statistics?

Answer 2

It’s when the relationship between two variables depends on a third variable (the moderator).

Question 3

What is the role of a moderator variable?

Answer 3

It changes the strength or direction of the association between a predictor and an outcome.

Question 4

In the heart attack diagnosis example, what moderates the relationship between symptoms and diagnosis accuracy?

Answer 4

Gender.

Question 5

Who is more likely to be misdiagnosed after heart attack symptoms?

Answer 5

Women.

Question 6

In the pain and swearing example, what moderates the relationship between fear of pain and experienced pain?

Answer 6

Swearing.

Question 7

What happens to the fear-pain relationship when people swear?

Answer 7

The association weakens—pain ratings drop.

Question 8

What are the three components in a moderation model?

Answer 8

Predictor (X), Moderator (W), and Outcome (Y).

Question 9

What is the moderation effect?

Answer 9

The interaction between the predictor and the moderator (X × W).

Question 10

What is a main effect?

Answer 10

The average effect of a predictor or moderator on the outcome.

Question 11

What is an interaction effect?

Answer 11

The combined effect of the predictor and moderator on the outcome.

Question 12

What does a stronger main effect of the predictor look like on a graph?

Answer 12

Steeper lines.

Question 13

What does a pure moderation effect look like?

Answer 13

Lines cross over—no main effects, only interaction.

Question 14

What JAMOVI module is used for moderation analysis?

Answer 14

“medmod”.

Question 15

What does “medmod” do?

Answer 15

Automatically creates a regression model testing main and interaction effects.

Question 16

What statistical distribution does “medmod” use?

Answer 16

Wald Z distribution.

Question 17

What is mediation in research?

Answer 17

Mediation explains the relationship between two variables through a third variable, which accounts for the association.

Question 18

When is mediation used?

Answer 18

When answering “Why” or “How” questions about relationships between variables.

Question 19

Is mediation causal?

Answer 19

No, mediation is not inherently causal; it is based on theoretical ordering of variables.

Question 20

What are the three main paths in a mediation model?

Answer 20

a path: Predictor → Mediator
b path: Mediator → Outcome
c’ path: Direct effect of Predictor → Outcome (controlling for mediator)

Question 21

What is the indirect effect in mediation?

Answer 21

The product of the a and b paths (a × b), representing the effect of the predictor on the outcome through the mediator.

Question 22

What is the direct effect in mediation?

Answer 22

The c’ path, which is the effect of the predictor on the outcome after accounting for the mediator.

Question 23

Why do cats purr? (Example 1)

Answer 23

Purring may communicate submission, which increases social survival—submission mediates the relationship.

Question 24

Why do cats purr? (Example 2)

Answer 24

Purring causes vibrations (25–50 Hz) that may promote healing—vibration mediates the relationship.

Question 25

What does the “medmod” module in JAMOVI do? What method should be used to calculate standard error in mediation?

Answer 25

It automates mediation analysis by creating regression models for each path and the indirect effect. The Bootstrap method, which resamples the data (e.g., 1000 times) for more accurate estimates.

Question 26

What are the four steps to report mediation results?

Answer 26

Describe the a path (predictor → mediator) Describe the b path (mediator → outcome) Describe the indirect effect (a × b) Describe the c' path (predictor → outcome, controlling for mediator)

Question 27

What indicates full mediation?

Answer 27

A significant indirect effect and a non-significant c' path.

Question 28

What indicates partial mediation?

Answer 28

A significant indirect effect and a significant c' path.

Question 29

How does ice cream make people happy (mediation example)?

Answer 29

Ice cream increases sugar/fat intake (a path), which increases happiness (b path). The indirect effect is significant, and c' path determines full or partial mediation.

Question 30

Why does food on a white plate seem less spicy?

Answer 30

Plate color affects expectations, which mediate perceived spiciness. If both indirect and c' paths are significant → partial mediation.

Question 31

What is moderation in statistics?

Answer 31

Moderation occurs when the relationship between two variables depends on a third variable.

Question 32

What are the three main patterns of moderation?

Answer 32

Magnify/Strengthen, Attenuate/Weaken, and Crossover.

Question 33

What does a Magnify/Strengthen pattern indicate?

Answer 33

A positive or negative association becomes more extreme depending on the moderator.

Question 34

What does an Attenuate/Weaken pattern indicate?

Answer 34

A positive or negative association becomes closer to zero depending on the moderator.

Question 35

What does a Crossover pattern indicate?

Answer 35

An overall null association hides significant positive and/or negative slopes depending on the moderator.

Question 36

Give an example of a Magnify/Strengthen moderation pattern.

Answer 36

People with high emotional eating eat more chocolate if they have positive attitudes toward chocolate.

Question 37

Give an example of an Attenuate/Weaken moderation pattern.

Answer 37

People high in depressive symptoms express less negative affect despite experiencing it, weakening the usual positive association.

Question 38

Give an example of a Crossover moderation pattern.

Answer 38

Relationship satisfaction and pain: higher satisfaction leads to less pain when holding a partner’s hand, but more pain when not holding anyone’s hand.

Question 39

What are simple slopes in moderation analysis?

Answer 39

They estimate the direction and strength of the relationship between variables at specific values of the moderator.

Question 40

How are simple slopes typically calculated for continuous moderators?

Answer 40

At low (−1 SD), average (mean), and high (+1 SD) values of the moderator.

Question 41

What does a significant moderation effect tell us?

Answer 41

That the relationship between variables differs depending on the moderator, but not which specific slopes are significant.

Question 42

Can all simple slopes be significant even if moderation is significant?

Answer 42

Yes, none, some, or all can be significant.

Question 43

What is a curvilinear effect in moderation?

Answer 43

When the relationship between a predictor and an outcome depends on the value of the predictor itself

Question 44

How is a curvilinear effect tested?

Answer 44

Using a regression model with the predictor and its squared term (e.g., Age and Age²).

Question 45

What does a positive curvilinear effect look like?

Answer 45

A U-shaped curve (e.g., happiness increases again at older ages).

Question 46

What does a negative curvilinear effect look like?

Answer 46

An inverted U-shape (e.g., productivity decreases at very high levels of happiness).