[fintstone]

You can do a "study" and get any conclusion you like once you decide it is ok to stray from actual data and start extrapolating based on some other factor. The reader of the study has to examine how the folks doing the "study" arrived at their numbers and if the reader finds it acceptable. As someone who used to do these studies for a living, I can tell you that there is no real way to determine with any confidence how many women in the U.S. were raped and did not report it (although if you want to hire me to do so, I can think of plenty of strategies where we can come up with a number that you can use...and justify by citing my study).