[Alex3917]

"I like your hypothesis about greater willingness to admit to "bad behavior" in the distant past than in the recent past. My intuition is that anonymity weakens that effect, but we'll have to measure to know for certain."

If you look at this report on the validity of self-reported drug use, it goes into the issue of how people are more likely to admit bad behavior that happened long ago. Anonymity probably does ameliorate the problem, but I'm guessing that it would still be significant.

http://archives.drugabuse.gov/pdf/monographs/monograph167/do...

[jseliger]

Great link—thanks for posting it—too few people cite the sources for their beliefs. In a related context, "Truth and consequences: using the bogus pipeline to examine sex differences in self-reported sexuality" discusses how "some of the sex differences in self-reports of sexuality are not due to actual sex differences in behavior, but rather to differences in reporting as a function of differential normative expectations for men and women": http://www.thefreelibrary.com/Truth+and+consequences%3a+usin... .