Using the 'medmod' module in Jamovi to test for mediation and moderation

[@tinettejayme4077]

How would you interpret results if mediation estimate outputs are all negative but p values are all significant?

[@nmny22]

@mikecrowson2462 why does the simple moderation plot have negative values on the x and y axis? is that interpretable?

[@mz555]

How does this compare to using PROCESS in SPSS? I did not see in the video any options to specify more complex models.

[@mikecrowson2462]

Hi there. What you see in this video are pretty simple models for sure. Go to Jamovi library. Under there, there is a package you can install called, jAMM (advanced mediation models). Once installed, you should have the capability of performing many more analyses akin to those in Process - including moderated mediation. I have it installed and it is under the medmod icon as jAMM GLM mediation model. It's very cool! Maybe I'll do a video on it in the future. Have fun with it!

[@alamer11]

@@mikecrowson2462 Yes. I saw it and played with it. It is cool to have such complexity in a simple way. However, unfortunately, lacking model fit indexes in the results is a serious flaw.

[@muraliyasvanth]

Sir, is it possible to use mediators and moderator like process macro

[@AnwarSadatShimul]

Hi Mike, thank you for the simple and clear tutorial. I have just been curious if Jamovi uses mean-centred values for the moderation. Thanks.

[@sczarche5149]

Hello sir, what does it mean if i have a significant positive direct effect between x and m, and x and y. However, i also have a significant negative indirect effect. Thanks

[@katefox4836]

Hi Mike, Just new to mediation and moderation - so I just had a question about my interpretation of the mediation analysis output. From looking at the path estimates, we can see the positive estimate between MGoal and SMint level, and again between SMint and Achievement - so would I interpret that to mean that to mean that when mastery goals increase, interest in the subject matter also increases, which in turn increases overall achievement? And we can see that the path between MGoal and Achievement is not significant in itself - is this what indicates that SMint is truly acting as a mediator in this scenario?

[@szczpk769]

HI, how long does it take to do medmod? It is possible that jamovie needs more than 3h to do any statistics?

[@szczpk769]

i think about bootstrap, 5000samples

[@mikecrowson2462]

Hi there. I just saw your note. I would not think it should take more than 3 hours to do any statistics, especially with only 5000 bootstrap resamples.

[@houdafaris4760]

Thank you, please could you explain us how to run a Tobit model in stata

[@adriannabocian8076]

simple and informative tutorial, thank you :)

[@Cjstistajh]

can I analyse moderated moderated-mediation in Jamovi

[@mikecrowson2462]

Hi Benny, assuming you have installed the jamm (Advanced mediation models module), you can go under MedMod and then GLM Mediation Model then you can test path models where there is moderation of paths. You can get conditional indirect effects. Unfortunately, it does not look like you can obtain the index of moderated mediation (IMM). The IMM can be computed using Hayes (2018) approach; however, you cannot generate a bootstrap confidence interval for the IMM for testing whether there is linear moderation of the indirect effect. Hopefully they add this one day. Cheers!

[@Cjstistajh]

@@mikecrowson2462 I see, thank you. So does that means if my model is a non recursive moderated moderated-mediation model, then it’s better that I just use Amos SEM? Because Hayes SPSS PROCESS cannot analyses non recursive mediation model?

[@annenolan3863]

why didn't you call the direct path c prime

[@mikecrowson2462]

It's a long time since I put this together. It was likely an oversight. The path from x-->y controlling for the mediator is generally treated as c-prime

[@yangzonggongga7376]

Hi Mike, this video is so useful. I'm confusing that can JAMOVI do mediated moderation analysis?

[@lightman2258]

Thank you very much 😁🌷

[@ermankilinc]

Thank you sir 🤓

[@muraliyasvanth]

Nice explanation