Hello,
I have a formative Construct with 4 Indicators. In Case of testing Multicollinearity it is recommended to check the bivariate correlation of the Indikators at first. I checked the indicators for bivariate Correlation (Pearson) in SPSS and found out, that two Indicators correlate completely as the correlation coefficient is 1. Does this implicate that I can delete one of this two Indicators at once without further testing of VIF and CI?
Another question: If there is a construct where all Indicators weights are not significant. how can i interpret this?
Thank you in advance,
Sten