I want to test if job type, personality type and gender of a person have an effect on how their effectiveness and hireability is perceived. So my independent variables are job A and job B, personality A and personality B, and male and female gender. My dependent variables are effectiveness and hireability.
I was planning on using MANOVA for my model but I've read something about a multivariate regression and I'm not exactly sure what the difference is or which one is more suitable. I'm also not sure how a multivariate regression would be calculated in SPSS?
Thank you for your help!
See multivariate regression in johnson and Wichern, applied multiariate statistical methods. Available at be z-lib , Google this. David Booth
Analysis of structural equations (path models and multiple regression)
Hello Annie,
If you have in mind a specific model, positing how the two sets of variables inter-relate, you'd likely be best served by going the SEM (structural equation model) route (with or without latent variables, as appropriate).
If you characterize your work as more an exploration, then either multivariate regression or canonical correlation could be helpful for you, as to attempt to explicate what's going on with your specific data set.
If you're thinking of something else entirely, then you'd have to elaborate your query a bit to help others to give more focused recommendations.
Good luck with your work.
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Statistics
- Statistical Software
- SPSS