Mediation

[nikolas7]

Hi all,

In my model i have consumer loyalty to mediate the relationship between quality and willingness to pay. However, thee is no direct effect between quality and willingness to pay. Using Preacher's and Hayes (2004,2008) method, it seems that a partial mediation effect exists. Can this be true though, as not direct effect exists?

Thank you.

regards,
Nikolas

[nikolas7]

Can anyone provide support on thi please?

Thank you.

Regards,
Nikolas

[Choukri]

Hi,
Could you explaine more, please
Greetings

[jmbecker]

Generally, the mediation literature says that full mediation exists if there is a significant indirect effect but no direct effect if the mediator is included in the model.
Partial mediation exists if there is both a significant indirect and a significant direct effect in the model with the mediator.

As you say that there is no direct effect, what makes you think that it is a partial mediation?

[nikolas7]

Yes, i think i put my question wrong, indeed it could not be a partial mediation only full mediation.
My main issue is, that in the full model i can not prove relationship between quality and Willingness to Pay, but when i am using mediation test i can see that exists.
My main question is, can i support that there is mediation relationship if i fail to prove any relationship in my full model?

Thank you

[jmbecker]

Sorry, I don't quite understand the problem yet. What is the "full model"? Could you describe the situations one by one?

[Choukri]

I think, in this situation, you should only focus on indirect effect (a*b) if it is significant or not, because you are not interested by direct effect.

Greetings

[nikolas7]

Thank you all.
Just to clarify here is what i mean full model and mediation model.
My full model contains quality, satisfaction, loyalty and willingness to pay (wtp). In that model there is no relationship between quality and wtp.
To test for mediation i first run a model with only quality and wtp. if that relationship is significant i add consumer loyalty.If quality to consumer loyalty is significant and consume loyalty to wtp is significant then mediation exists. This according to Hair 2013

[jmbecker]

Why are you excluding Satisfaction from your mediation model?
Don't run these step by step mediation analysis, which focus only on the mediation relationship. Just analyse your complete model and assess the indirect and direct effect.

[donnadba]

I read the question by nikolas7 and the response from jmbecker on mediation. I have been looking for the reference on the criteria for full and partial mediation not using the VAF or Sobel test. If Dr. Becker can provide it I would be most appreciative. Thanks.

[jmbecker]

The best source for a general overview of a mediation model assessment process is:
Zhao, X., Lynch, J. G., & Chen, Q. (2010). Reconsidering Baron and Kenny: Myths and truths about mediation analysis. Journal of Consumer Research, 37, 197–206.

Other helpfull sources are:
Hayes, A. F. (2013). Introduction to mediation, moderation, and conditional process analysis: A regression-based approach. Guilford Press.
Preacher, K. J., & Hayes, A. F. (2004). SPSS and SAS procedures for estimating indiect effects in simple mediation models. Behavior Research Methods, Instruments, & Computers, 36(4), 717-731.
Preacher, K. J., & Hayes, A. F. (2008). Asymptotic and resampling strategies for assessing and comparing indirect effects in multiple mediator models. Behavior Research Methods, 40, 879–891.

[donnadba]

Thank you so much for posting these helpful resources. in the Zhao article they refer to "complementary" mediation instead of partial mediation. Would you use "partial" when writing an article at this point in time?

[jmbecker]

With reference to Zhao et al. I would go for the more specific complementary or competetive mediation (which are kind of sub categories of partial mediation).

[Hengkov]

See also:

https://global.oup.com/academic/product ... id&lang=en&

http://www.guilford.com/books/Doing-Sta ... 1462508150

Best regards,