via Instapundit (I know, I know), we have a several-day-old-article about a new poll counting the number of Baghdad residents Saddam Hussein murdered. The estimate they come up with is 61,000, in Baghdad alone.
This was higher than previous estimates, but it does not really surprise me. I never doubted him to be a thug and a murderer and whatever invective I could come up with–though every time you’re confronted with the numbers again you wonder how the Iraqi people could possibly be worse off when all this is over, no matter how much we screw it up.
But what I’m really interested in, for the purposes of this post, is the methodology:
“The survey obtained Monday, which the polling firm planned to release on Tuesday, asked 1,178 Baghdad residents in August and September whether a member of their household had been executed by Saddam’s regime. According to Gallup, 6.6 percent said yes.
The polling firm took metropolitan Baghdad’s population — 6.39 million — and average household size — 6.9 people — to calculate that 61,000 people were executed during Saddam’s rule. Past estimates were in the low tens of thousands. Most are believed to have been buried in mass graves.”
I’ve never heard of this casualty-count-by-poll before. Do we have any statisticians or social scientists reading this? Is this a decent method of counting casualties? How does it differ from the use of press reports or anecdotal evidence, or mass graves, or the various other methods that are used?
I have no idea if it’s even remotely accurate. But if it is, we should consider using it to figure out how many civilians (or for that matter soldiers) were killed in the U.S. invasion, and its aftermath.
(continued, and please read the whole thing before commenting.)
Read more