I understand the benefits of
metadata collection. I want to know what school children who are failing have in common so that by targeting that, maybe I can have more children succeed. Is it the school, the teacher, socioeconomics, health, language, etc. If you look at one child's academic results, it's difficult to understand why that child is failing. But if you look at dozens or hundreds or thousands of children's results, patterns emerge that can be used to pinpoint the source the problem and suggest ways to improve teaching methods.
But I also understand that metadata collection and analysis is unpopular in some circles. I'm not unsympathetic to some of their concerns.