The McClatchy Washington bureau, which has done some remarkable reporting over the last few years, caught a lot of flak this week over one of its stories about casualties in Iraq. In response, the bureau posted a story about
how it derived those figures.You may agree or disagree with its conclusions, but it's worth reading because of the tone and thorough look it brought to its own work--no defensiveness, no half-baked apology, just reporting on reporting.
What the McClatchy story reminds us is how careful we have to be with language. Are casualties up or down and is the surge to blame/deserving of credit?
When the White House issues its report next week on the effects of the troop increase in Iraq, we should keep in mind the need to be extra careful in how we describe these numbers.
(
UPDATE: According to the Washington Times, there will be no written report:
A senior military officer said there will be no written presentation to the president on security and stability in Iraq. "There is no report. It is an assessment provided by them by testimony," the officer said.
The only hard copy will be Gen. Petraeus' opening statement to Congress, scheduled for Monday, along with any charts he will use in explaining the results of the troop surge in Baghdad over the past several months. That testimony will follow the meeting of the president, Gen. Petraeus and Mr. Crocker at the White House.)
Here are some of McClatchy's numbers:
Month Hostile Non-Hostile Total
Aug 2007 56 28
84
July 2007 67 12
79
June 2007 93 8
101May 2007 120 6
126Aug 2006 58 7
65July 2006 38 5
43 June 2006 57 4
61May 2006 57 12
69Aug 2005 75 10
85 July 2005 45 9
54June 2005 69 9
78May 2005 67 13
80Aug 2004 55 11
66 July 2004 44 10
54June 2004 37 5
42May 2004 62 18
80Aug 2003 16 19
35July 2003 28 20
48June 2003 18 12
30 May 2003 8 29
37One of the issues comes in what statistics you're comparing. If you talk about spring vs. summer, U.S. casualties are down. But if you're comparing August 2007 and August 2006, for example, they're up.
The Washington Post had an extraordinary story yesterday about how the U.S. military
categorizes casualties. How are we to reach consensus with this kind of parsing going on?
The intelligence community has its own problems with military calculations. Intelligence analysts computing aggregate levels of violence against civilians for the NIE puzzled over how the military designated attacks as combat, sectarian or criminal, according to one senior intelligence official in Washington. "If a bullet went through the back of the head, it's sectarian," the official said. "If it went through the front, it's criminal."
"Depending on which numbers you pick," he said, "you get a different outcome." Analysts found "trend lines . . . going in different directions" compared with previous years, when numbers in different categories varied widely but trended in the same direction. "It began to look like spaghetti."
UPDATE 9/9
The AP looks at the statistics morasse.
Iraq debate is sea of statistics
By RICHARD LARDNER
WASHINGTON - In vertical bars of blue, green, gray and red, a briefing chart prepared by the Defense Intelligence Agency says what Gen. David Petraeus won't.
Insurgent attacks against Iraqi civilians, their security forces and U.S. troops remain high, according to the document obtained by The Associated Press. It is a conclusion that the well-regarded Army officer who is the top U.S. commander in Iraq is expected to try to counter when he and Ryan Crocker, the U.S. ambassador in Baghdad, testify before Congress on Monday and Tuesday.
More than four years into a conflict initially thought to be a cakewalk, the war has become a battle of statistics, graphs and conflicting assessments of progress in a country of more than 27 million people.
The defense intelligence chart makes the point, with figures from Petraeus' command in Baghdad, the Multinational Force-Iraq. Congressional auditors used the same numbers to conclude that Iraqis are as unsafe now as they were six months ago; the Bush administration and military officials also using those figures say that finding is flawed.
With so much depending on how the statistics are collected and interpreted, policymakers in Washington are confused.
Several of us, colleagues at a newspaper years ago but now mostly scattered to the winds geographically and choices of other careers, have been engaged in an interesting e-mail discussion provoked by the E&P article by Thomas "Dennie" Williams about staff cutbacks and investigative reporting. Topics include the failure of newsroom people to answer e-mail or phone messages (absolutely not a new issue but one that seems to have worsened in recent years, some people think), the damage to reporting and editing from the cutbacks, the viability of print at all and a bunch of other matters.
