Everybody is drawing the conclusion from the NPD study that the RIAA's legislation threat actually forced people to delete files. Here's NPD's summary. I think this a big leap. There are several methodological problems with this study.
First, the files that were deleted weren't monitored at the time of download - so there's no way of knowing whether they came from. Second, how much did users trust NPD not to essentially as a cop for the RIAA, and inform on them? Third, NPD doesn't say which P2P services it was monitoring. Now user adoption can ebb pretty fast - maybe they were monitoring the decaying ones.
Here's another telling statistic about sample selection bias:
"...Eighty percent of the consumers who deleted files had fewer than 50 files saved; just 10 percent had more than 200 files."
Did any of these journalists even bother to read NPD's summary? It's pretty obvious from reading it that this is a seriously shaky survey.