Slashdot posted a story about an Identity Theft study conducted by interviewing people convicted of identity theft. A more detailed post to which Slashdot referred has more details.

The supposed conclusion drawn by Heith Copes (University of Alabama at Birmingham) and Lynne Vieraitis (University of Texas at Austin) is:

"Despite public perceptions of identity theft being a high-tech,
computer driven crime, it is rather mundane and requires few technical
skills. Identity thieves do not need to know how to hack into large,
secure databases. They can simply dig through garbage or pay insiders
for information. No particular group has a monopoly on the skills
needed to be a capable identity thief."

The flaw in their work of course is that they only interviewed thieves who were caught!! In order to really draw meaningful conclusions about identity theft they need to interview thieves who did not get caught.
Of course that increases the difficulty dramatically.

It reminds me of a story I heard years ago from my friend Joe. One night he was walking down a narrow street that had only one street light. Under it was a drunk who seemed to be looking for something. My friend Joe went up to him and asked if he could help. The drunk said, "Sure, I lost my keys and I'm looking for them." My friend asked the drunk, "Where did you lose them?" The drunk responded, "Over there." My friend asked, "Then why are you looking over here?" The drunk answered, "Well it's dark over there. The light is over here."