Scribd
Upload
283 views8 pages

1.moderation and Mediation Analysis Using Process Macro

This document discusses moderation and mediation analysis using the Process Macro. It defines moderation as a change in the strength of the relationship between two variables due to a third moderator variable. Mediation is defined as an indirect effect of the independent variable on the dependent variable through a mediator variable. The document explains direct, indirect, and total effects and how to calculate partially and fully standardized indirect effects. It also discusses ratios of indirect to total/direct effects and how to calculate the proportion of variance explained by an indirect effect.

Uploaded by

Kelly Kavita
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
283 views8 pages

1.moderation and Mediation Analysis Using Process Macro

This document discusses moderation and mediation analysis using the Process Macro. It defines moderation as a change in the strength of the relationship between two variables due to a third moderator variable. Mediation is defined as an indirect effect of the independent variable on the dependent variable through a mediator variable. The document explains direct, indirect, and total effects and how to calculate partially and fully standardized indirect effects. It also discusses ratios of indirect to total/direct effects and how to calculate the proportion of variance explained by an indirect effect.

Uploaded by

Kelly Kavita
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 8

Moderation and Mediation Analysis

using Process Macro


Sanjay Singh, PhD
www.sanjaysingh.org

Sanjay Singh, PhD | www.sanjaysingh.org


What is Moderation
Refers to change in strength of relationship between two variables due to a
third variable called moderator. A higher designation
leads to higher salary.
But Gender may have
Moderator Gender
important moderating
influence as females are
generally paid less .

Stress leads to
depression. But the
X Y relationship becomes
moderated by Social
support.
Designation Salary

Sanjay Singh, PhD | www.sanjaysingh.org


What is Mediation

• Mediation is measured via indirect effect of X on Y


via mediator M.

• The significance level of indirect effect is measured


by “Sobel Test” which is essentially a t-test that
measures whether indirect is significantly different
than 0 (i.e., no indirect effect). Its same as t-test in
linear regression or Wald’s test in logistic regression.

Sanjay Singh, PhD | www.sanjaysingh.org


X Y
c

c = Total Effect

c' = Direct Effect

axb = indirect effect

Mediator

a b

X Y
c'

Sanjay Singh, PhD | www.sanjaysingh.org


Direct, Indirect and Total Effect
• Total Effect = Effect of X on Y when M is not included in model
• Direct Effect = Effect of X on Y when M is included in model

• Indirect Effect = Combined Effect of X on M (a) & M on Y (b)= axb


(where ‘a’ and ‘b’ are unstandardized beta weights)

Sanjay Singh, PhD | www.sanjaysingh.org


Partially standardized indirect effect vs Completely
standardized indirect effect
• Since indirect effect is expressed as ‘axb’ which is unstandardized beta so its useful to
convert them into standard beta weights
• Partially standardized indirect effect refers to standardization of indirect effect by
dividing it with the standard deviation of dependent variable (Y)
Partially standardized indirect effect = axb/Standard deviation of DV

• Partially standardized indirect effect does not take into account the variation in IV and
Mediator variable hence it is called ‘partially standardized’.

• To fully standardize the indirect effect we can multiply the Partially standardized indirect
effect with standard deviation of predictor (Preacher & Hays, 2008)
Thus, Fully standardized indirect effect = (axb/Standard deviation of DV) x Standard deviation of IV
This is also called “Index of Mediation”

Sanjay Singh, PhD | www.sanjaysingh.org


Ratios of Indirect Effect

• Ratio of indirect to total effect of X on Y = ab/Total Effect


This tells what % of total effect is indirect effect

• Ratio of indirect to direct effect of X on Y = ab/Direct Effect


This tells what % of direct effect is indirect effect

• These ratios can be misleading as they may go above 1 or can be


negative. They should be used only when sample size is 500 to 5000
respectively.
Sanjay Singh, PhD | www.sanjaysingh.org
Proportion of variance explained by Indirect Effect
• R2indirect = R2mediator – (R2mediator and predictor - R2predictor)
or R2M = R2Y,M – (R2Y, MX - R2Y,X)

Where,
R2M = Proportion of variance explained by indirect effect
R2Y,M = Proportion of variance explained by mediator
R2Y, MX = Proportion of variance explained by mediator and predictor
R2Y,X = Proportion of variance explained by predictor

Sanjay Singh, PhD | www.sanjaysingh.org

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy