The sample size is too small to do any useful feedback and humans are too complex to be mapped/binned in the way interviews attempt to do. Not only that but th

The sample size is too small to do any useful feedback and humans are too complex to be mapped/binned in the way interviews attempt to do. Not only that but things that happen after hiring may have a bigger influence on the success/failure of the hire than the factors the interview is looking for. E.g. who you end up working with, internal politics, the specifics of the job. Kahenman's work on the illusion of validity comes to mind. It kind of boils down to our hiring process must be right because we're doing it.