And even if after all that, they still do? Its not going to be a statistically significant proportion of users who participate in studies to skew the results.

And even if after all that, they still do? Its not going to be a statistically significant proportion of users who participate in studies to skew the results. That's beautiful, and absolutely true. When I was younger, I remember answering surveys for random offers. I was caught a few times by different surveys for inconsistencies - although I wasn't told explicitly. To cope with this, I drafted up several identities and thought through a backstory to each character - personality, background, etc, which made it much easier to remember details. Depending on what the survey was about, I just picked a profile I felt a researcher would be looking for. Wasn't perfect, but had a hugely high success rate. I doubt very many people will go to my length though. When you start trying to passively filter people, and hit someone like me - well, we see it as a challenge... for better or for worse. An interesting note: I used one of these profiles far more than the others. As a result, I remember walking home one day near midnight, and thinking "Geeze, I'm 29 already... one more year until I'm 30 - and I still haven't done anything I wanted to do." It didn't strike me until a minute or so later, that I wasn't anywhere near 29, and it wasn't actually my birthday.