jAMM GLM Mediation Model - continuous mediator with multiple

General help and assistance with jamovi. Bug reports can be made at our issues page: https://github.com/jamovi/jamovi/issues . (If you're unsure feel free to discuss it here)
Post Reply
BenH
Posts: 7
Joined: Tue Sep 21, 2021 5:26 pm

jAMM GLM Mediation Model - continuous mediator with multiple

Post by BenH »

We ran a study where we asked participants to respond on a Likert scale with their strength of preference for three types of policies. We now want to test whether policy support mediates the relationship between two other variables and whether the type of policy matters.

What is the best way of setting up jAMM GLM Mediation Models (or any other Jamovi module) to allow us to test a conditional mediation where the moderator is categorical (policy level) and the mediator is continuous (policy preference)?
User avatar
mcfanda@gmail.com
Posts: 457
Joined: Thu Mar 23, 2017 9:24 pm

Re: jAMM GLM Mediation Model - continuous mediator with mult

Post by mcfanda@gmail.com »

Hi
do you have three variables preference, one for each policy, or do you have one variable that refers to all three policies together? In other words, is policy type between subject or within subjects?
BenH
Posts: 7
Joined: Tue Sep 21, 2021 5:26 pm

Re: jAMM GLM Mediation Model - continuous mediator with mult

Post by BenH »

Hi,
Thank you very much for your response.
Each participant responded with their strength of approval for each of the three policy types so we have three policy preference/approval variables for each participant (so they are within subject variables).
Your advice would be very much appreciated.
Many Thanks
Ben
User avatar
mcfanda@gmail.com
Posts: 457
Joined: Thu Mar 23, 2017 9:24 pm

Re: jAMM GLM Mediation Model - continuous mediator with mult

Post by mcfanda@gmail.com »

HI
as regards testing the mediation you can put the three preference variables as mediators, the independent variable as covariate. With this model, you get three mediated effects, one for each policy. This model, however, cannot test the difference in the mediated effects. So, you can evaluate the mediated effect but not compare them statistically.

If you really need to test the differences among mediated effects, you need to use the path analysis module (PATHj). With PATHj you can estimate the same model you obtain in jAMM, but it allows testing for equality (and thus differences) between coefficients within a model. It's the equality constraints method (if you need to look it up).
BenH
Posts: 7
Joined: Tue Sep 21, 2021 5:26 pm

Re: jAMM GLM Mediation Model - continuous mediator with mult

Post by BenH »

Hi
Thank you so much for your response.
I would really like to use the path analysis module but unfortunately it does not appear in the Jamovi library on my desktop app (1.6.23.0) and it is also missing from https://www.jamovi.org/library.html . Is there anywhere else that I can download it from?
Many Thanks
Ben
User avatar
MAgojam
Posts: 421
Joined: Thu Jun 08, 2017 2:33 pm
Location: Parma (Italy)

Re: jAMM GLM Mediation Model - continuous mediator with mult

Post by MAgojam »

Hi, BenH.

The current 0.7.0 version of Marcello Gallucci's pathj module is not compatible with the solid 1.6.23 version of jamovi you are using.
Therefore it is not available in the desktop app library.
In order to install the pathj module you will need to upgrade your jamovi to version 1.8.0 or higher.
This is the link to the line in the module 0000.yaml file:
https://github.com/pathj/pathj/blob/mas ... 0.yaml#L37
I think Marcello can confirm.

Cheers,
Maurizio
User avatar
mcfanda@gmail.com
Posts: 457
Joined: Thu Mar 23, 2017 9:24 pm

Re: jAMM GLM Mediation Model - continuous mediator with mult

Post by mcfanda@gmail.com »

sure, as Maurizio said, you can get pathj in version >1.8.0. jamovi version 2.0.0 works fine and gives you all the modules available
BenH
Posts: 7
Joined: Tue Sep 21, 2021 5:26 pm

Re: jAMM GLM Mediation Model - continuous mediator with mult

Post by BenH »

Thank you very much for your support
Post Reply