Thanks for sharing your knowledge Prof. Crowson. It is absolutely enlightening. Could you let us know how to run this moderated mediation model type 7 using MPlus syntax? And how if the model is multilevel 2-2-1?
Hello Mike. Great job. Please, is it possible to input all your variables into the Process Macro to run the analysis? I have 1 IV, 3 DVS, 2 Mediators and 2 Moderators
Thanks for the video! If one was using this model within a feasibility study, thus wanted to look more specifically towards effect sizes, which numbers would be reported for which pathways? It looks like there's r-sq for the IV to respective mediation paths, and from IV to DV, but what about the b pathways from mediation(s) to DV?
Thank you for all the explanation; it was very clear. I would like to give you a "Thanks," but I can´t see the option. You should activate it to let us thank you.
Really clear video, thank you so much! I'm doing a parallel moderated mediation analysis, does it matter if my X and W are both categorical (two levels)?
Thanks for your video! And do you know what if the moderator in two mediation path is in different place? Like the moderator is in x-m for mediation path1, and in m-y for mediation path2?
I am running model 76 (moderated-mediation model), and I wonder if all interaction is non-significant and the test(s) of highest order unconditional interaction(s) is non-significant as well, but the conditional direct effect (s) of X on Y is significant, can I say that the moderator has effect on the direct relation between X and Y?
hey mike... i have a question. what's the model number for running a model with two serial mediators, but ONE moderator between X and M1? It'd be so helpful for me if you could revert. thank you!
Hello sir, Would you please advice me what's the model number for 2 serial mediators and a moderator betwen IV and DV? Sir, it will greatly help me. Please!
Thank you so much for all your tutorials, very helpful. Is it possible to extend Model 7 further to 4 mediators? I tried ist and I can calculate in SPSS even with 4 Mediators, I am just not sure if there exists a problem by doing it I don´t realize yet...
Hi Aisha, unfortunately his macro is only set up for one X variable (which is a limitation). If you want to use multiple X's, you'd need to use another program. You might try SEM, but it would require some programming. Another option is a free-ware called jamovi (download at www.jamovi.org/download.html). If you add the 'medmod' module, there is an option for glm mediation model. Under there you can specify a moderated mediation model with multiple X's, mediators, and moderators. I haven't done a full video on that, but it is pretty slick. It is somewhat tricky to build your model, so you'd have to play with it a bit. Also, I don't believe it currently will give you the index of moderated mediation. However, it will provide conditional indirect effects. I hope this is helpful!
I really appreciate for your videos! but I have a question to ask. If X is categorical variable which is manipulated into two groups, which analysis should I use? Can I still use the same method as you showed after coding two groups into 0 and 1? Thank you so much!
Hello Mike Crowson, thank you for this very clear explanation. Funny how you can write your master thesis using all new things from youtube ;). I have a question left and I'll try to explain where I get lost. So, my independent variable is a three leveled (so 2 dummies) variable. As learned from the powerpoint, one can say that something is a (non-)significant positive OR negative predictor, falling at the mean of.. In my first sub model, my moderator is significant positive predictor, falling at the mean of my independent variable. But this variable exist out of two dummies. My output also says 'product key terms' in which int_1 = X1 x moderator and int_2 = X2 x moderator. Should I work with those two instead of the 'single moderator'? I hope you understand what I mean, thank you in advance!
Dear Mike,
Bless you for all your help to researchers around the world. Thank you for the clear explanations and making science accessible to all!
You are very welcome, Maya. Thanks for visiting! Have a great weekend :)
Wonderfully done, underrated video.
Thanks for sharing your knowledge Prof. Crowson. It is absolutely enlightening. Could you let us know how to run this moderated mediation model type 7 using MPlus syntax? And how if the model is multilevel 2-2-1?
Hello Mike. Great job. Please, is it possible to input all your variables into the Process Macro to run the analysis? I have 1 IV, 3 DVS, 2 Mediators and 2 Moderators
Thanks for the video! If one was using this model within a feasibility study, thus wanted to look more specifically towards effect sizes, which numbers would be reported for which pathways? It looks like there's r-sq for the IV to respective mediation paths, and from IV to DV, but what about the b pathways from mediation(s) to DV?
Thank you for all the explanation; it was very clear. I would like to give you a "Thanks," but I can´t see the option. You should activate it to let us thank you.
Really clear video, thank you so much! I'm doing a parallel moderated mediation analysis, does it matter if my X and W are both categorical (two levels)?
Thanks for your video! And do you know what if the moderator in two mediation path is in different place? Like the moderator is in x-m for mediation path1, and in m-y for mediation path2?
I am running model 76 (moderated-mediation model), and I wonder if all interaction is non-significant and the test(s) of highest order unconditional interaction(s) is non-significant as well, but the conditional direct effect (s) of X on Y is significant, can I say that the moderator has effect on the direct relation between X and Y?
Is it necessary to draw covariance between the manifest variables (exogenous in my case)? I have 12 exogenous in my model.
hey mike... i have a question. what's the model number for running a model with two serial mediators, but ONE moderator between X and M1? It'd be so helpful for me if you could revert. thank you!
Model 83 has X->M1-M2->Y, with W moderating the path from X->M1
Hello sir,
Would you please advice me what's the model number for 2 serial mediators and a moderator betwen IV and DV? Sir, it will greatly help me. Please!
Thank you so much for all your tutorials, very helpful. Is it possible to extend Model 7 further to 4 mediators? I tried ist and I can calculate in SPSS even with 4 Mediators, I am just not sure if there exists a problem by doing it I don´t realize yet...
It is very helpful.. Thank you.. can we use Hayes' Process macro with two IV categorical variables (X)?.
Hi Aisha, unfortunately his macro is only set up for one X variable (which is a limitation). If you want to use multiple X's, you'd need to use another program. You might try SEM, but it would require some programming. Another option is a free-ware called jamovi (download at www.jamovi.org/download.html). If you add the 'medmod' module, there is an option for glm mediation model. Under there you can specify a moderated mediation model with multiple X's, mediators, and moderators. I haven't done a full video on that, but it is pretty slick. It is somewhat tricky to build your model, so you'd have to play with it a bit. Also, I don't believe it currently will give you the index of moderated mediation. However, it will provide conditional indirect effects. I hope this is helpful!
I really appreciate for your videos! but I have a question to ask. If X is categorical variable which is manipulated into two groups, which analysis should I use? Can I still use the same method as you showed after coding two groups into 0 and 1? Thank you so much!
how do I do the model with my variables?
When M-Y is nonline relationships. How to do
Thank you so much
Hello Mike Crowson, thank you for this very clear explanation. Funny how you can write your master thesis using all new things from youtube ;).
I have a question left and I'll try to explain where I get lost.
So, my independent variable is a three leveled (so 2 dummies) variable.
As learned from the powerpoint, one can say that something is a (non-)significant positive OR negative predictor, falling at the mean of..
In my first sub model, my moderator is significant positive predictor, falling at the mean of my independent variable. But this variable exist out of two dummies. My output also says 'product key terms' in which int_1 = X1 x moderator and int_2 = X2 x moderator. Should I work with those two instead of the 'single moderator'?
I hope you understand what I mean, thank you in advance!