8. Hayes Process Macro Model 6 - Serial Mediation

[@heatherkartsounis2571]

Thank you so much for making this video! It's really helped me with my thesis!

[@researchwithfawad]

Glad it was helpful!

[@ASMM1981EGY]

I'm Egyptian Muslim Genetic Anthropologist and I'm really proud of you being a great Muslim researcher and scientist too ❤🙏

[@researchwithfawad]

Jazak Allah Khair. Thank you.

[@sidrahasib499]

thnx alot dr its clear now was strugling with alot with serial mediation

[@researchwithfawad]

Thanks. I am glad you liked it.

[@mimihasan7291]

Thank you so much for making this video. MashaAllah

[@researchwithfawad]

Pleasure. I am glad you liked it.

[@julieyang3016]

Very helpful🙏thank you so much for sharing this!

[@researchwithfawad]

Glad it was helpful!

[@cailawrance8369]

Awesome! Thanks for sharing!

[@researchwithfawad]

Thanks. I am glad you liked it

[@Dalenatton]

Thank you so much for sharing. I wonder if there is a way to compare the indirect effect of the original serial mediation model (commitment --> assurance) with the indirect effect of the reverse serial mediation model (assurance --> commitment), assuming that both models are significant, with one indirect effect larger than the other. Thank you!

[@researchwithfawad]

Hi. You can design two different model and then make the comparison.

[@laila29]

Question: what if your indirect of one mediator is insignificant (so ind2) but your indirect effect through both mediators (ind3) is significant and ind1 is also significant? Is there still serial mediation?

[@researchwithfawad]

No.

[@Alhamzah_F_Abbas]

Very good explanation. can Process Macro be an alternative powerful tool for mediation and moderation instead of AMOS?

[@researchwithfawad]

Thanks. I am glad you like it. Not a complete alternative but yes it is used.

[@burcuakpinar2339]

Thank you so much for the video. However, I have 2 predictors, 2 mediators, and 3 outcome variables. If I understood correctly, it seems impossible for me to put both of my predictors into the same analysis. Not in the SPSS I guess. Do you have any suggestions? Or is it okay to run the mediation analysis 2 times by putting the predictors separately?

[@researchwithfawad]

Pleasure. I am glad you like it. I recommend you to use SmartPLS or AMOS.

[@amatiagolbodaghi4200]

Hello, thank you for this video. I have a clarifying question. If your X is a Professional Coaching and the variable is categorical meaning there is a group coached and a group that is not coached (meaning 2 categories not 3) how would you run this analysis?

[@researchwithfawad]

Thanks for watching. Please watch kzbin.info/www/bejne/in-xfJKiaM-DZtk

[@lauravanalphen7244]

Hi there! Thank you so much for this video! I do have a question; my direct and total effect are significant, but when looking at the indirect effects, only one is significant. Do I still have serial mediation? Because ind effect 3 (so the complete one) is also not significant?

[@researchwithfawad]

Thanks for watching. If one of the indirect effect is significant, this mean there is mediation through that particular mediator, the other indirect effects are insignificant, so no mediation through those.

[@Yash_k_28]

In this model, can we analyse three mediators, i.e. A -- B -- C -- D -- E, A --E (A leads to B, B leads to C, C leads to D, D leads to E, A leads to E)

[@researchwithfawad]

Yes, if my memory serves me right, Model 6 allows upto 6 mediators.

[@abcfakename]

Hello there. Fantastic description of the serial mediation analysis. Quick question, why is the R-Square value for the Total Effect model (.3086) smaller than the R-Square smaller than the R-Square value for the model summary (.4737)? Doesn't the total effect model incorporate all the direct and indirect effects? Shouldn't it explain more variance in performance (OP) than just the direct effects in the model summary?

[@researchwithfawad]

Thanks for watching. Total effect model is without the inclusion of the mediators.

[@kirangupta7362]

Hello Sir, in your notes, in the example all indirect paths are significant. What if Ind1 is insignificant but Ind2 & Ind3 are significant? Can it still be called as serial mediation? Then how to report the results? Also, will it be complementary or competitive in such case?

[@researchwithfawad]

No, that won't be serial if the whole path is not significant.

[@EdenTan-v3e]

Good day, sir. I would like to ask if there is a serial mediation effect, when direct effect of X on Y is not significant, but my indirect effects and total effect are all significant. I have another situation which is when my direct and total effect are significant, but one of my indirect effect is not significant while others are significant, is this still considered as serial mediation (partial or full mediation)?

[@researchwithfawad]

Yes there is mediation in both cases.

[@ayeshafarooq-f3e]

If both uper and lower bootstrap confidence interval are negative then what will be the interpretation??

[@researchwithfawad]

Significant results as there is no zero in between the two values.

[@ayeshafarooq-f3e]

Thankyouu

[@droffilcx]

Good day! This video is very helpful. Thank you for creating this. I just have a question. What does it mean if my 1st mediator's mediation is not significant in the overall model, but when I try to separately analyze the mediation of this 1st mediator, it is significant? For context, my serial mediation was found to be significant, it's just that the 1st mediator's mediation was not. Thank you!!!

[@researchwithfawad]

Thanks for your interest. A mediator may be significant in individual relationship but in serial it may be insignificant, as it is not establishing the chain of relationship.

[@anasifti203]

Is it possible in this serial mediation to have full mediation with individual mediators then a partial mediation with both mediators?

[@researchwithfawad]

There are a set of models that can be analyzed using process macro. You may search Google for Hayes Process Macro Models for the different models.

[@deekshasingh2257]

What if indirect effect (IND1) is not significant but IND2 and IND3 is significant?

[@researchwithfawad]

There is no mediation.

[@deekshasingh2257]

​@@researchwithfawad is there no serial mediation or simple mediation? As when I am using model 4 separately simple mediation is there for both mediators

[@kamalbadar5770]

Do we use this model if we want to test serial-moderated-mediation?

[@researchwithfawad]

No. You may need to check the different model that can be tested using Hayes Process Macro. You may search Model in Hayes Process Macro in Google.

[@kellyberden7394]

Hi, i tried this, and all my indirect effects are significant (t-value >1.96 and no zero between upper and lower bounds). But I have a total effect off 1.1670 (.000), T-value bigger than 1.96: but there is a zero between upper and lower bound,.Also with the direct effect, i have .5941(0.050), T-value bigger than 1.96, but a zero between upper and lower bound... How do I interpret this, do I have a Full mediation then? And how do i put this in the table? Because you dont mention that in your tabel only the effect and p-value... Is it neccessary to have no zero in between upper and lower? Why is that? What does it mean? I also have no zero between B2, but do have a T--value bigger than 1.96 and p

[@kellyberden7394]

I also put gender and age in the model as controlvariables, but dont do anything with is

[@researchwithfawad]

If both the direct and indirect effects are significant, it's partial mediation. Otherwise, if the indirect effect is significant but direct is insignificant, it's full. As for 0 in between, please check again, a 0 in between LLCI and ULCI means insignificant results.

[@m.sswetha3165]

Happy Day sir Incase if there the p- value is not significance When Culture is culture is the only variable what does it mean

[@researchwithfawad]

p value over 0.05 mean no significant results.

[@kalemguy]

Can we do serial mediation for unbalanced panel data using Process Macro? and can we use financial secondary panel data for this analysis?

[@researchwithfawad]

I am sorry but i do not have any experience with financial data.

[@elaineleezieyi1832]

May I know how to calculate the t value for serial mediation?

[@researchwithfawad]

To do so, Divide Total Indirect Effect by SE. You will find that in Indirect effect(s) of X on Y: In the example it is .3402/.0577

[@nikedejong5724]

Is the bootstrap used here Percentile bootstrap?

[@researchwithfawad]

By default, PROCESS versions 3 and later produces bootstrap confidence intervals using the percentile method.

[@destaneysauls796]

Can I ask how you added the process options to the drop down menu? I've always just written the syntax

[@researchwithfawad]

Thanks for your comment. Here is the link that will help to install the Process Macro www.processmacro.org/uploads/3/4/8/5/34858384/dialoginstall.pdf

[@destaneysauls796]

@@researchwithfawad Wonderful! Thank you for the help - and I appreciate the quick response!

[@MashbbabM.Almashouf]

can you please provide the data you use for model 6

[@researchwithfawad]

For Sample Data, please visit researchwithfawad.com/index.php/download-research-resources/

[@ezgiguneyuygun8474]

Hi! Are a2 and b1 essential paths for serial mediation model?

[@researchwithfawad]

Yes, the path have to linked if you want to establish the serial mediation.

[@sidrahasib499]

can you tell us how to perform serial mediation on smart pls 4

[@researchwithfawad]

You can add the variable and link them just as you do in a conceptual framework.