boutinb
commented
4 years ago @vandenman: do you know this package?
Closed cbflynn1970 closed 9 months ago
boutinb
commented
4 years ago @vandenman: do you know this package?
EJWagenmakers
commented
4 years ago This is something to discuss with @vankesteren
vankesteren
commented
4 years ago Thank you for your suggestion, we'll keep an eye on this!
Just FYI, we're generally not fans of the obscure model 17, model 4 etc. user interface of PROCESS. Which type of model do you want to run? We might be able to provide a nice user interface for that model, specifically.
Also, I was wondering what you mean by "has limits on the number of mediators one can use"? How many mediators do you want to use? See below a JASP mediation model with 16 mediators, that seems like a lot to me already?
cbflynn1970
commented
4 years ago I get the dislike of the models; I don’t particularly care for them myself, however, PROCESS does provide a variety of moderated-mediated models that are useful. Having the ability to run models with moderators before and after the mediator would be nice, but primarily before the mediation variable. I guess you can create moderated variables in the data set and just run those, but it would be nice if you could just specify multiple moderators in the model and where they fall in the paths. I think just having the ability to do multiple moderators with the current capabilities of the mediation would probably be sufficient.
In terms of PROCESS this would be similar to models 1-3, 5, 7, 8, 9, 14, 21, etc.
Yes, 16 mediators is a lot; I’ll never run that many. So I guess what I would really like is an easy way to add moderation to the mediation model.
And while I’m asking, what’s the possibility of adding an HLM module and a visual model builder to the SEM module (a la AMOS; I shy away from writing code, even in Llavaan)?
Just a compliment to all of you, I really like JASP - It’s my go-to software for stats and I really appreciate that it is available freely. I prefer to do most of my work on my Mac but getting a good stats package for Mac was hard. I tried R but it’s a little over my head and interest. I just want a package where I can input my data and run it and JASP has been great. I use it with my students for class projects as well. So kudos to all of you for the great work.
Dr. Brian Flynn Assistant Professor of Management Coggin College of Business University of North Florida 904-620-1118 brian.flynn@unf.edumailto:brian.flynn@unf.edu
On Apr 15, 2020, at 11:04 AM, Erik-Jan van Kesteren notifications@github.com<mailto:notifications@github.com> wrote:
Thank you for your suggestion, we'll keep an eye on this!
Just FYI, we're generally not fans of the obscure model 17, model 4 etc. user interface of PROCESS. Which type of model do you want to run? We might be able to provide a nice user interface for that model, specifically.
Also, I was wondering what you mean by "has limits on the number of mediators one can use"? How many mediators do you want to use? See below a JASP mediation model with 16 mediators, that seems like a lot to me already?
— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHubhttps://nam01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fjasp-stats%2Fjasp-issues%2Fissues%2F675%23issuecomment-614094157&data=02%7C01%7Cbrian.flynn%40unf.edu%7C397881495c4340c657d808d7e14e3bcf%7Cdf29b2fa8929482f9dbb60ff4df224c4%7C1%7C0%7C637225598442000020&sdata=PH4DD269%2FKkTiLqAqb8x2J4rXbS7OSS3NB2whswn3fU%3D&reserved=0, or unsubscribehttps://nam01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fnotifications%2Funsubscribe-auth%2FANBGUPJNJBERQ2CA3W5LH23RMXEGDANCNFSM4MHX3P3A&data=02%7C01%7Cbrian.flynn%40unf.edu%7C397881495c4340c657d808d7e14e3bcf%7Cdf29b2fa8929482f9dbb60ff4df224c4%7C1%7C0%7C637225598442000020&sdata=7h9L7kuVebebiYAl7OibbM3p7brTkKAbr74Esul5Oro%3D&reserved=0.
lwnn25
commented
9 months ago Dear all,
I find that JASP can use PROCESS code in R console directly.
First, download the PROCESS R file on https://haskayne.ucalgary.ca/CCRAM/resource-hub.
Second, copy all the codes in the PROCESS R file and paste into R console in JASP. Run the code and wait for the information - "PROCESS is now ready for use."
Third, write the code for mediation or moderation analysis and run it.
Here is an example for mediation analysis:
process(data = data, y = "maths", x = "age", m ="intelligence", model = 4)
Just replace names of y, x, m and the model with your data and analysis.
Enjoy it.
EJWagenmakers
commented
9 months ago Thanks! The upcoming version of JASP will have a PROCESS model with extensive functionality and GUI! EJ
Enhancement: Andrew Hayes has published that he will be porting his moderation-mediation package PROCESS to R sometime in April 2020. I know that JASP uses a lot of R sub-processes. Any consideration for porting PROCESS for R into the JASP environment?
Purpose: Provide a robust way to run various models of moderation and mediation Is your feature request related to a problem? Please describe. The current mediation is great, but has limits on the number of mediators one can use, and the variety of models one can use.
Describe the solution you'd like Porting PROCESS would provide a more robust mediation/moderation solution
Describe alternatives you've considered I use PROCESS in SPSS now, but would like to get out of the proprietary environment and be able to do everything in JASP.