__
1) it was voluntary so there is some self selecting group of people who choose to play on a computer and answer personal questions.
2) people lie on surveys all the time - just try to find accurate numbers on # of sex partners or the like
3) I don't understand the 'statistical analysis' that determines the birthday answer to the drugs answer
4) Conceivably,honorable athletes could lie and say they dope in order to get more dollars to WADA to catch the cheats
5) carbon testing is pretty accurate as I understand, and that found a 5% doping rate, right? I may be remembering that wrong. This number is so large that you have to question it.
You're right on the limits of physical testing, but that is the best thing we got right now. This sort of social science y polling means little, to my mind.