Mplus Mediated Moderation

In this video I show and explain how to do a mediated moderation analysis in Mplus using a floodlight plot or J-N plot. I think in the video I call it moderated mediation by accident...

[sajjadulhassan]

Thank you sir for this great video.
Nowadays researchers are more inclined towards multilevel modeling, so if you please make some video tutorials for 2-1-1 with level one moderator, and level two moderator and similar for 2-2-1, 1-2-2, 1-2-1, 2-1-2 models with 1st as well as 2nd level moderators. Thanks

[javeda]

Please add more videos on the causal mediation on Mplus too

[ibrahimcolak]

James, thanks for your great videos. My question is what if our model includes mediator and moderator as latent variables. Can I create interaction by multiplying latent variables? How can I construct such a model?

[lozw]

Hi James, really helpful video. When the moderated mediation is with a logistic regression (as my DV is binary), I am just wondering when reading the indirect effect, whether I look at the total indirect effect (and associated plot Total Natural IE) or at the Odds ratio of binary Y indirect effect (with the associated plot called Odds Ratio, Total Natural IE). Any help would be appreciated! Thanks :)

[TheH]

Hi James, thank you so much for sharing your knowledge for the SEM!! I am wondering that can you teach us how to calculate the indirect effects after being moderated, please? Look forward for your new video!!

[ItsSnagret]

Hi James, I'm wondering if you would know how to do with with latent variables? I'm running a model with one IV, one Med, one Mod, and one DV, all latent variables. I'm trying to get a plot that makes sense, but it's hard to understand how to do it.

[lilyabrera]

Thank you for this video! I ran the model. However, it ignores the standardize request. Any idea why this is happening and how to solve it? I am really needing the standardized results. Thank you!

[affanasif]

I am also desperately waiting for your videos on multilevel modeling by using MPlus

[YaqiGao]

Hi James, thanks for the video! I was trying to examine moderation, indirect effect and conditional indirect effect in one model. When I simply did the path modeling, the moderation effect is significant (p value of the interaction term is lower than .05). When I conducted bootstrap, the p value of the interaction term became larger than .05, and the bias-corrected CI of this interaction term included 0. But the bias-corrected CI for indirect effects and conditional effects were excluding 0. So in this case can I say there is a significant moderation effect? Thank you very much!

[md.mostafizurrahman]

Thank you so much. I gathered almost all my knowledge from you through your precious videos. I have 2 IVs, 2 Mediators, 1 Moderator, and 1 DV. I want to run a moderated mediation model (the moderator is in between IVs and Mediators. Do I need to run the model for two separate IVs? I want to find indirect effects with (+ -) 1 stand. dev. with bootstrapping LLCI and ULCI. How do I find the M+ codes for this....?

[umerubaca]

Hello James, your videos are amazing and you really did a great job in making my basic concepts clear. Well, may i know if there is any video available that can help me regarding codes and reporting of results ( for example: how to interpret results regarding acceptable values or goodness of fit of model). Acutyally, i am conducting a diary study based on multilevel structural equation modeling using mplus, having 1 IV (4 items), 2 Mediators (parallel, 10, 1 items), 1 DV (4 items) all at day level and 1 moderator on IV-Mediators path at between level, where all variables are latent.

[researchassistant]

Thank you for your videos. They are really helpful! I am trying to figure out how to run a moderated mediation model with two DVs. All variables are continuos. However, I have yet to find syntax that accommodates for two DVs. Any thoughts about this?

[ki-wonhaan]

Thank you so much. I see that this method replicates using MODEL CONSTRAINT. I had one question: How can we find the value -0.92 in the plot? I found it by going through iterative process changing the intervals but I was wondering if there is a faster way. Does (1 0) after the IV indicate the direction of the CI bands?

[ammaarkidwai]

Hi James! Thanks for the video. I'm wondering why there isn't a section in your code for Analysis? Including Type, Estimator, etc...? For example, I cannot run an input file without first noting that Type = Random due to the interaction term being created. If you could clarify, that would be so helpful!

[DrFurb]

Life saver. Been trying to figure this out for 2 weeks! But how many IVS can you include in that model? And what is the (1 0) you said not to worry about for now?

I also notice you only include one IV (Feedback) in your syntax. Why have you not included Custrej in that syntax chain?

[miaoinperth]

Hi James, the title of the video is mediated moderation but the title of the syntax is moderated mediation. Is it just an error in the title of the video?

[mortazazare]

Great video! I do have a question though. Wouldn't it be necessary to have mean centering for moderation analysis?

[tarikuayana]

Dear James, thank you for this and other useful videos on mplus. I have five independent variables (latent factors), one mediator, one moderator, and three outcome variables. I am struggling to test LMS moderated mediation model. May you suggest to me some resources or syntax with similar model? Thank you in advance!

[NotToday]

Would running a multiple group analysis be a form of moderation?

[munibasaleem]

would the same process apply for a categorical moderator with 2 levels?