Dear Holly Davies

These links might be resourceful, have a look:

1. https://jasp-stats.org/2018/07/03/how-to-perform-structural-equation-modeling-in-jasp/

2. https://www.researchgate.net/post/Has_anyone_used_JAMOVI_or_JASP_for_SEM_analysis

3. https://forum.cogsci.nl/discussion/5852/different-results-when-using-mediation-and-sem-module-in-jasp

4. https://inis.iaea.org/collection/NCLCollectionStore/_Public/42/040/42040081.pdf

Kind Regards

Qamar Ul Islam

Holly, you might want to specify whether you want the comparison about them computationally (so JAPS's sem is a front-end to lavaan, so all those comparisons apply) or as how well it fits within a teaching curriculum (and you mentioned price, so obviously there is a difference there)?

Personally, I am generally an Mplus user, and I really like that program (https://www.statmodel.com/). It is more expensive than any R package (obviously), but the support you have access to with your license is unrivaled, in my opinion. I don't know of an R package that gives you personal responses to your questions/support requests from folks like Linda and Bengt Muthen, themselves.

However, R is free and open source. So if cost is a significant barrier, then R might be your better bet. Many of my colleagues use R regularly, as well.

I think the most important point as a methodologist, though, is to not marry yourself to one particular program. I have found that there are times that a particular model will not converge in one stats package, but will be estimable in another package. So, having multiple tools in your data analytic toolbox may help you down the line.

Best of luck!

~Sarah