From the actual second report:

"Gunshot wounds caused 56 percent of violent deaths, with car bombs and other explosions causing 14 percent, according to the survey results. Of the violent deaths that occurred after the invasion, 31 percent were caused by coalition forces or airstrikes, the respondents said."




As to Teribus' statement:

"I would tend to go on what has been verified, as opposed to what has been projected."

Statistical projections are used all the time in the sciences (not just social science) when phenomena cannot actually be verified. These methods are almost never questioned in other sciences. This study impacts rather strongly upon politics, so methods that are lauded in other sciences are in this case derided.

Can we at least agree that "what has been verified" is undoubtedly an underestimate?

This then begs the question of underestimated by how much? And how should one go about answering that question?