That sounds pretty bad as you present it. I have seen a state uni do a similar sounding kind of employee survey -- the point of which was not to violate individ

That sounds pretty bad as you present it. I have seen a state uni do a similar sounding kind of employee survey -- the point of which was not to violate individual privacy but to produce an aggregated survey to present to health insurance companies bidding on providing employees that benefit. I imagine there's a pretty huge incumbent bias there; the current provider has exact data on claims and they aren't sharing with the competition. And especially in the insurance industry, there's a concern about adverse selection and asymmetric information that I imagine could be cured or at least reduced with additional data.