The Johns Hopkins study that concludes that approximately 600,000 Iraqis have been killed since the US invasion in March 2003 has naturally put the war's supporters (which includes the US media) on the defensive. The only question has been: how best to discredit it?
I was bitterly amused to read the San Francisco Chronicle's attempt. First, the title of the article--the first original article the Chronicle has devoted to the topic--is "Critics say 600,000 Iraqi dead doesn't tally," thus framing the entire issue from the viewpoint of critics rather than highlighting the study's horrific conclusions. The subheading is "But pollsters defend methods used in Johns Hopkins study"; we'll see just who these "pollsters" are later.
So who are the critics? Following one general and vaguely disparaging quote ("If the number of civilian casualties cited in the report is anywhere near the true number, it calls into question the legitimacy of the whole campaign") from a source who we later find actually supports the study's methodology, the article cites the following renowned experts on statistical sampling, in this order: George Bush, US General George Casey, Iraqi government spokesman Ali Dabbagh, and Anthony Cordesman of the Center for Strategic and International Studies. So the article starts out by citing four separate critics of the study--none of whom have the necessary statistical background to comment meaningfully on the study's methods or conclusions, and all of whom have clear political motivations for rejecting it out of hand. As you'd expect, they offer no concrete criticisms at all--nothing but vacuous attacks and dismissals. Bush's statement that the study's methodology was "pretty well discredited" previously is a particularly mindless howler, not that that's any surprise considering the source (the full quote, for your amusement: "this report is one -- they put it out before, it was pretty well -- the methodology was pretty well discredited.")
Finally, in paragraph 13, we hear one person without any obvious statistics credentials (Loren Thompson of the Lexington Institute, the source of the previously-cited general quote) defend the study. We then have to wait until paragraph 18 to hear from sources with expertise in the relevant subject area, namely John Zogby, the founder of a well-known polling agency; Ronald Waldman, an epidemiologist at Columbia University who previously worked at the Centers for Disease Control and Prevention, who calls the survey method "tried and true" and says the survey is "the best estimate of mortality we have" (paragraph 20); and Frank Harrell Jr., chairman of the biostatistics department at Vanderbilt University, who praised the study for its "rigorous, well-justified analysis of the data" (paragraph 21).
So the only people mentioned in the article who actually have the necessary expertise to evalute the study are cited deep in the bowels of the article--and dead last of those giving direct commentary on the study. And although only one of the three is associated with a polling agency while two of them are academics with strong and clearly relevant credentials, the three are summarized together in the subheading of the article as "pollsters." Lazy readers likely won't even make it this far, and those who only read the article title will have no idea that 67% of the "pollsters" are in fact qualified academics who are experts in the field in question.
(As an aside, it's interesting to look at the sources for many of these quotes as well. The Chronicle is a strange mix of a low-budget hometown newspaper and a major metropolitan daily, and it shows in this article; at least three of the quotes are sourced to the AP and the Washington Post, showing that the author just read articles from other outlets and didn't even bother to contact the sources herself. You'd think that having the Stanford and UC Berkeley campuses within shouting distance would encourage her to seek out some experts of her own--but that would only have undercut the purpose of the article.)
So the Chronicle's first original story on this critically important study is framed from the viewpoint of critics who are unqualified to comment on its methodology and who are clearly motivated to attack it whatever the facts might be. It cites those critics prominently for 12 paragraphs before allowing any commentary that's supportive of the study's methodology. And it relegates the comments of the only truly qualified commentators--those speaking from expertise rather than from a predetermined political agenda--to the deep nether regions of the article. Yet I have no doubt that the author would defend the article as being balanced, since it cites a roughly equal number of supporters and detractors. Ask yourself how often you see this approach with similar reports about mortality in Darfur, or the Congo, or anywhere else where statistical sampling methods are used...that is, when the results are politically expedient or neutral, instead of damning as in this case.
And all of this disingenuous chicanery is meant to mask the simple, horrible truth that the US invasion of Iraq has resulted in the deaths of approximately 600,000 Iraqis--to our eternal shame.
Comments