I'm very new to this and have 1 Independent Variable ( Working from home) and its effect on 3 dependent variables ( loneliness, creativity and resilience) . Can someone please confirm I will use multivariate regression? Can I run this in Jamovi ? And I need to work out the sample size using G Power?
Hello Neeta,
1. You certainly can use multivariate regression. Whether it makes sense to do so, as opposed to running individual regression models, depends on your specific research question(s) (as well as the usual assumptions).
2. It's not in the default Jamovi menus, but Jamovi has a SEM add-in module, in which you could declare and estimate a path model from your lone IV to three DVs.
3. For a priori sample size determination, you could use the "F tests/Manova-global effects" option as a guide (three response variables). The other approach is the "F tests/Generic F test." Either way, the challenge is to identify a defensible effect size that makes sense given the goals of your research and the specific variables involved.
Good luck with your work.
@Neeta @David has given you excellent advice. If you find it difficult to run path analysis model in JAMOVI (taking one IV and three DVs) as it works with Lavaan for SEM and therefore is command based. You can also use Smart-PLS to test your model. It works well on categorical/binary data well as it seems your IV (work from home) might be a categorical variable. Wishing you best.
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Sampling Design
- Sample Size