Interesting discussion. I would add that the next step after convergent validity is for it to be externally validated and for the researchers to perform intraclass correlation testing. the take home is that the measurement needs to demonstrate a meaningful signal between different groups.