Casual Models

Question 1

Causality

Answer 1

• If x causes y then changes in x produce changes in y
• Causes must happen before effect
• Time it takes for x to change y is irrelevant
• Effects of x on y may vary in context, x changes y in some situations but not others
• Cause and effect must happen in spatial context or through chain of events

Question 2

Bidirectional/Reciprocal Relationship

Answer 2

Two variables influence one another

Question 3

Complete Mediation

Answer 3

The only way for X to influence Y is through Z

Only used when you are not able to clearly identify another mediator other than Z.

Question 4

Direct Causal Relationship

Answer 4

A cause has direct effect/impact on an outcome variable.

x -> y

Question 5

First-Order Moderated Variable

Answer 5

Directly moderating effects of X on Y

Question 6

Mediated Moderation

Answer 6

Adding a mediator to a moderated relationship (asses for partial or complete).

• Taking a philosophy course (x) and improving logical reasoning (y)
• Moderator: someone desires to be there and therefore pays more attention and learns more
• Mediator: attendance, desire to take course also effects attendance which then also effects ability to learn
• Partial mediation: desire to take course may not always impact attendance but will still impact learning

Question 7

Indirect Causal Relationship

Answer 7

Variable influences another indirectly though it’s impact on a third variable.

x -> z -> y

Question 8

Mediator Variable

Answer 8

the intermediating variable (z) the determinant (x) works through to cause change in outcome variable (y).

x -> Z -> y

Question 9

Moderated Casual Relationship

Answer 9

Casual relationship is dependent on a third variable (z).

Moderator variable impacts the casual relationship, so when visually shown can only even point to the arrow connecting two variables.

Question 10

Moderated Mediation

Answer 10

Strength of (casual) path signifying mediation varies due to third variable.

When visually shown the moderator impacts the relationship between a determinate and a moderated variable.
Can be applied to both complete and partial mediation.

Question 11

Moderated Moderation

Answer 11

Amount at which the moderated variable impacts the effect of one variable onto another is also determined by a fourth variable.

When visually shown moderated moderator points towards arrow of first moderator.

Question 12

Moderator Variable

Answer 12

differing Z variable (male or female) produces different causal relationship between x (wearing revealing clothes) and y (likelihood to receive harassment of some kind).

Question 13

Partial Mediation

Answer 13

In addition to the mediated effects through Z, X also has an independent effect on Y.

When used you are saying there is some other effect other than Z that also influences Y. If you can articulate it then it is partial, but you do not always want to indicate it due to overcomplication.

Question 14

Second-Order Moderated Variable

Answer 14

Moderating effects of the first moderator

Question 15

Spurious Relationship

Answer 15

Variables are related because they share a common cause, not because they are casually related to one another.

x y

Question 16

Unanalysed Relationships

Answer 16

Two variables are related (casual connection or spurious) but theorist won’t specify why due to overcomplication drawing away from main purpose of what is actually being discussed.

Relationships could be infinite but one must ignore some for clear understanding to he had.
Every single possible relationship does not need to be listed.

Question 17

Predictive Models

Answer 17

Is variable A related to variation in variable B?

Does not care how A may cause B, only that it can’t predict B.
- certain personality traits (A) can predict mental disorders (B)