I I have a model (attached). All loadings are higer than 0,7 except of hma4 and bag1. HMA factor's AVE is 0,56 and BAG factor's AVE is 0,53. Composite reliability of these factors are above 0,80. Should I delete these (hma4 and bag1) variables to upgrade reliability and validity?
You also have ak1 and ak3 below 0.7. Although ak 3 is nearly 0.7. I suggest deleting ak1, hma4 and bag1. Run a few simulations with random data to test the reliability and validity.
Another approach would be to combine (ak1 and ak3), (bag 1 and bag 6) and (hma 4 and hma 1) as a single predictors and recalculate. This method may strengthen your predictive ability.
Good luck. Hope this helps.
Mr. Christopher Smith. I think combining variables will solve the problem. I never thought about this solution. Thank you very much for your help
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Statistics
- Probability
- Probability Theory
- Reliability