Mediation: Bootstrapping the indirect effect

Questions about the implementation and application of the PLS-SEM method, that are not related to the usage of the SmartPLS software.
Post Reply
nicefreak
PLS User
Posts: 11
Joined: Thu Nov 04, 2010 1:48 pm
Real name and title:

Mediation: Bootstrapping the indirect effect

Post by nicefreak »

Hello everyone!

When trying to test for mediation, I stumbled about two questions for the "bootstrapping the indirect effect"-approach:

1) As far as I know bootstrapping the indirect effect is usually done by subtracting the direct effect from the total effect for each bootstrap sample. What is the reason not just to multiply a*b for each bootstrap sample?

2) If one has found a full mediation (direct effect turns insignificant when mediator is introduced), is there a way to use the "bootstrapping the indirect effect" approach?

Thank you for your help!
jmbecker
SmartPLS Developer
Posts: 1282
Joined: Tue Mar 28, 2006 11:09 am
Real name and title: Dr. Jan-Michael Becker

Post by jmbecker »

1) Actually both approaches should give you the same results.

2) As far as I have understand Preacher and Hayes it is simply

IF a*b is significant (in bootstrapping) we have at least partial mediation.
IF, in addition, c (the direct link) becomes not significant, we have full mediation.
nicefreak
PLS User
Posts: 11
Joined: Thu Nov 04, 2010 1:48 pm
Real name and title:

Post by nicefreak »

Thanks a lot for the carification,

Martin
ruchi
PLS Expert User
Posts: 75
Joined: Wed Dec 22, 2010 3:09 pm
Real name and title:

Post by ruchi »

Hi

From reading all the post, what i understood is like below
Here I will represent two models-MODEL 1( without mediating variable) and MODEL 2(with Mediating Variable)

Model 1

X---> Y this path was insignificant.

Even in model 1 if the path is insignificant between X and Y , I can have indirect effect . So i build Model 2 with mediating effect

Model 2
X----> Y
X--->M --->Y

When i did bootstrapping on MODEL 2 to get significance , i got
X----> Y This was again-insignificant
X--->M Significant
M--->Y Significant

Now here I can say that there is FULL mediation.

Please give me ur suggestions or let me know if i m wrong.

Thanks
Ruchi
User avatar
Diogenes
PLS Super-Expert
Posts: 899
Joined: Sat Oct 15, 2005 5:13 pm
Real name and title:
Location: São Paulo - BRAZIL
Contact:

Post by Diogenes »

Hi,

In fact the path from X --> Y could be significant,

But when we add the mediator, that coefficients will decrease.


“A variable functions as a mediator when it meets the following conditions: (a) variations in levels of the independent variable significantly account for variations in the presumed mediator (i.e., Path a), (b) variations in the mediator significantly account for variations in the dependent variable (i.e., Path b), and (c) when Paths a and b are controlled, a previously significant relation between the independent and dependent variables is no longer significant, with the strongest demonstration of mediation occurring when Path c is zero. In regard to the last condition we may envisage a continuum. When Path c is reduced to zero, we have strong evidence for a single, dominant mediator. (p.1176)”
BARON, R. M.; KENNY, D. A. The Moderator-Mediator Variable Distinction in Social Psychological Research: conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology, v.51, n.6, p.1173-1182, 1986. Available at: <http://www.public.asu.edu/~davidpm/clas ... /Baron.pdf>.

Best regards,

Bido
ruchi
PLS Expert User
Posts: 75
Joined: Wed Dec 22, 2010 3:09 pm
Real name and title:

Post by ruchi »

Dear Sir

I am really thankful to you for clearing my so many doubts.
It is really helping me in my modelling.

In the quotes mentioned , I am confused for the third(c) criterion.
(a) variations in levels of the independent variable significantly account for variations in the presumed mediator (i.e., Path (b) variations in the mediator significantly account for variations in the dependent variable (i.e., Path b),
(c) when Paths a and b are controlled, a previously significant relation between the independent and dependent variables is no longer significant, with the strongest demonstration of mediation occurring when Path c is zero. In regard to the last condition we may envisage a continuum. When Path c is reduced to zero, we have strong evidence for a single, dominant mediator. (p.1176)”

In this,

It means that even initially my path of X-->Y ( with No mediation) is insignificant , I can go ahead to model with mediation (X-->M--->Y). Again here

X----> Y This was again-insignificant
X--->M Significant ( meeting the criterion a )
M--->Y Significant ( meeting the criterion b )

Sir, does it mean that there is FULL MEDIATION or NO MEDIATION (as X----> Y is again insignificant)

Thanks
Regards
Ruchi
User avatar
Diogenes
PLS Super-Expert
Posts: 899
Joined: Sat Oct 15, 2005 5:13 pm
Real name and title:
Location: São Paulo - BRAZIL
Contact:

Post by Diogenes »

Hi,

X --> Y (insignificant = it means that there is no direct effect, all effect of X on Y is indirect or mediated).

Best regards,

Bido
ruchi
PLS Expert User
Posts: 75
Joined: Wed Dec 22, 2010 3:09 pm
Real name and title:

Post by ruchi »

Sir, Thanks a lot for clearing all my doubts

Regards
Ruchi
ruchi
PLS Expert User
Posts: 75
Joined: Wed Dec 22, 2010 3:09 pm
Real name and title:

Mediation

Post by ruchi »

Sir,

In the case of my model, the direct path X--> Y is significant in the model (where there is no mediator)

X---->y signifcicant
M-->Y significant


Then , a mediator is assumed in the model connected as below

X---->M----->Y
X---->Y

In this model. the path X---->M is insignificant; M----->Y is significant and X--> Y is also insignificant

what does this indicate??

regards
Ruchi
User avatar
Hengkov
PLS Super-Expert
Posts: 1599
Joined: Sun Apr 24, 2011 10:13 am
Real name and title: Hengky Latan
Location: AMQ, Indonesia
Contact:

Post by Hengkov »

Hi,

M is no mediator.

Best Regards,
Hengky
ruchi
PLS Expert User
Posts: 75
Joined: Wed Dec 22, 2010 3:09 pm
Real name and title:

Post by ruchi »

Ya, M is not acting as mediator

but how can the relation of X->Y becoming insignificant can be explained?
User avatar
Hengkov
PLS Super-Expert
Posts: 1599
Joined: Sun Apr 24, 2011 10:13 am
Real name and title: Hengky Latan
Location: AMQ, Indonesia
Contact:

Post by Hengkov »

Hi,

Check collinearity problem among X and M to Y.

Best Regards,
Hengky
konglong
PLS Junior User
Posts: 2
Joined: Wed Apr 03, 2013 6:33 am
Real name and title:

Post by konglong »

reading all the post, what i understood is like below
Here I will represent two models-MODEL 1( without mediating variable) and MODE



_________________________
aion kinah|Diablo 3 Gold kaufen|Diablo 3 Gold||Guild Wars gold|
bwilson
PLS Expert User
Posts: 41
Joined: Wed Apr 02, 2008 7:11 am
Real name and title:
Contact:

Post by bwilson »

Diogenes wrote:
BARON, R. M.; KENNY, D. A. The Moderator-Mediator Variable Distinction in Social Psychological Research: conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology, v.51, n.6, p.1173-1182, 1986. Available at:


Bido's response is great
You should see:
Wilson, Bradley; Henseler, Jörg (2006). The mediating role of relationship quality impacting sponsorship effects on perceived economic outcomes. ANZMAC Conference, Brisbane, December 4-6.

Henseler, Jörg; Wilson, Bradley; Westberg, Kate (2011). Managers' perceptions of the impact of sport sponsorship on brand equity. Sport Marketing Quarterly, 20 (1), 7-21.

All the best,
Brad
Bradley Wilson. Ph.D.
Senior Lecturer in Advertising.
RMIT University.
School of Media and Communication.
GPO Box 2476V
Location. 9.5.20
Melbourne. Victoria.
Australia.

SEE FOR PUBLICATIONS
www.rmit.edu.au/staff/bradleywilson
Post Reply