11 Flashcards
(24 cards)
What does mediation refer to in research?
Mediation refers to a situation when the relationship between a predictor variable and an outcome variable can be explained by their relationship to a third variable (the mediator).
Mediators help to elucidate how or why a predictor influences an outcome.
What are the components of mediation?
The components of mediation include:
* Predictor
* Outcome
* Mediator
Example: Predictor (heat), Outcome (crime rate), Mediator (discomfort)
What is the Baron & Kenny (1986) method for testing mediation?
The Baron & Kenny method involves three linear models:
1. Predicting the outcome from the predictor (total: c)
2. Predicting the mediator from the predictor (1st indirect: a)
3. Predicting the outcome from both the predictor and the mediator (2nd indirect: b)
This method checks if mediation conditions are met.
What are the four conditions of mediation according to Baron & Kenny?
The four conditions are:
* The predictor must significantly predict the outcome (model 1: Is c significant?)
* The predictor must significantly predict the mediator (model 2: Is a significant?)
* The mediator must significantly predict the outcome, controlling for the predictor (model 3: Is b significant?)
* The predictor must predict the outcome less strongly in model 3 than in model 1.
These conditions ensure that the mediation effect is properly established.
What is the difference between full mediation and partial mediation?
In full mediation, the direct effect of the predictor on the outcome becomes nonsignificant when the mediator is included. In partial mediation, the direct effect is reduced but still significant when the mediator is included.
This distinction helps to understand the strength of the mediation effect.
What does the indirect effect in mediation represent?
The indirect effect represents the joint effect size of both paths (a*b), where ‘a’ is the effect of the predictor on the mediator and ‘b’ is the effect of the mediator on the outcome.
The sign of the indirect effect should match the sign of the direct effect.
True or False: Mediation analysis can determine causation.
False
Mediation analysis is a causal model, and if the causal order of concepts is not established, it proves nothing.
What is bootstrapping in mediation analysis?
Bootstrapping is a statistical procedure that resamples a single dataset to create many simulated samples, providing more reliable estimates of standard error and confidence intervals.
It is particularly useful when data is skewed or when sample size is low.
What does a confidence interval (CI) indicate in mediation analysis?
A confidence interval indicates the range within which 95% of the resampled estimates fall. If both limits are positive or both are negative, the indirect path is significant.
A CI containing zero indicates that the effect is not significant.
What is the significance of the predictor variable in mediation?
The predictor variable (X) is essential as it influences both the mediator (M) and the outcome (Y), establishing the mediation pathway.
Understanding the role of the predictor helps in developing effective interventions.
Fill in the blank: The total effect in mediation is calculated as _______.
Direct effect + Indirect effect
This relationship helps in understanding how much of the effect is mediated.
What is a mediator’s role in a mediation model?
A mediator explains the relationship between the predictor and the outcome by accounting for some of the variance in the outcome.
This highlights the importance of the mediator in understanding the causal mechanisms.
What is the significance of the indirect effect in mediation analysis?
The indirect effect is significant if the confidence interval does not include 0.
In this case, the indirect effect a*b = 0.005, 95% CI [0.002, 0.08] indicates significance.
What tool was used for the mediation analysis?
Hayes (2022) Process tool in SPSS.
This tool allows for running mediation analyses with bootstrapping methods.
What were the results of the total effect before including the mediator?
c = 0.04, p < .001.
This indicates a significant total effect before considering the mediator.
What were the results of the direct effect after including the mediator?
c’ = 0.03, p < .001.
This shows that the direct effect remained significant even after including the mediator.
What is mediation?
Mediation refers to a situation when the relationship between two variables can be explained by their relationship to a third variable (the mediator).
It is a causal model that requires theory-led testing.
What method helps understand the mechanics behind mediation?
The Baron & Kenny (1986) method.
This method outlines the conditions necessary for establishing mediation.
What does the Process tool allow in mediation analysis?
Running a mediation analysis with bootstrapping method.
Bootstrapping provides a way to estimate the indirect effect and its confidence intervals.
What is the significance of p-values in mediation analysis?
p < .001 indicates a statistically significant result.
Common thresholds for significance are p < .05, p < .01, and p < .001.
Fill in the blank: The bootstrap estimate of the indirect effect is _______.
0.005, 95% CI [0.00, 0.01].
What does it mean if the mediation is full?
It means the direct effect is no longer significant when the mediator is included.
Full mediation indicates that the mediator fully accounts for the relationship between the independent and dependent variables.
What does it mean if the mediation is partial?
It means the direct effect remains significant when the mediator is included.
Partial mediation indicates that the mediator accounts for part of the relationship but not all.
What are unstandardized coefficients?
Unstandardized coefficients are raw scores that indicate the change in the dependent variable for a one-unit change in the independent variable.
They are used in regression analysis to interpret the direct effects.
