Hypothesis: People will lie on internal surveys more often if they feel either nothing will be done or an honest response could be used against them. I wouldn't

Hypothesis: People will lie on internal surveys more often if they feel either nothing will be done or an honest response could be used against them. I wouldn't put too much stock in the results of those surveys. Anecdotal support for hypothesis: Have been told firsthand by several people that they lie on these types of surveys and don't think results are truly anonymous. I'd love to see a broad study of this across companies, conducted by a third-party organization with no affiliation to any company studied. In general, I'd estimate anything below 4 (on a 1-to-5 scale) on those surveys could indicate a problem. What if "3/neutral" really means, "This actually sucks but I'm not willing to tell you that because I don't trust you"? I doubt there are many internal surveys showing "tons of unhappy people" unless employees a.) have no fear of company reprisal, for whatever reason; or b.) have entirely given up on the company and are in open rebellion.