One construct is family self-efficacy
Other is job self-efficacy...
both are Independent (exogenous) variables of my research. The pilot data show high correlation (0.8) between two...is is possible that I combine them and use as one construct?
Unless there is theoretical support and accepted by experts please don't combine constructs. If you still want to combine then go ahead for your convenience.
There is no statistical reason why you can't combine the two. I would begin by examining the reliability of each scale separately, and then the reliability for the combined scale.
If you are worried about multi-colinearity in regression, this will affect the ability to interpret the two separate b's associated with this pair of variables, but it will not impact their total effect on the dependent variable. This would allow you to enter them into the equation as a pair, for control purposes, even it is difficult to separate out their two effects.
- Badges
- Science topic
- Similar topics
- Medicine
- Physiology
- Biosignals
- Biomedical Signal Processing
- Statistical Analysis