Bootstrap of LV correlations
Posted: Sat Aug 23, 2008 10:56 am
If we have the standard error of the LV correlations, we could compute
- t-values and significances
- Interval of confidence. It's a way to show that the measurement model has discriminant validity ( r = 1 is out of the I.C.).
Best regards.
Bido
- t-values and significances
- Interval of confidence. It's a way to show that the measurement model has discriminant validity ( r = 1 is out of the I.C.).
Best regards.
Bido