> But in reality probably only .00000001% of facebook users probably quit because of privacy reasons, which is probably a drop in a bucket compare to the amount

> But in reality probably only .00000001% of facebook users probably quit because of privacy reasons, which is probably a drop in a bucket compare to the amount of new users they get everyday. Normalization is a tricky job. Your assumption is that the entire set of users were aware of privacy problems with Facebook. Out of the universal set of Facebook users, there would be two subsets - people aware of privacy problems and their ill effects and people who don't. You should check "how many people quit (or significantly reduced activity on) Facebook, once they found out about privacy issues and their effects" and then report the percentage.