Thursday, June 17, 2004

Poll-axe

Grim headlines sprout atop the story of the coalition-commissioned poll of 1,093 Iraqi adults in six cities — Baghdad, Basra, Mosul, Diwaniyah, Hillah and Baquba — taken May 14-23. "Disturbing" (Fox News), "Gloomy" (Business News). It is disturbing and gloomy. The numbers are way too high for my liking and for the safety of our troops. But is it really the stark proof of failure that the AP, reporting with undue malice, paints it to be?

The AP isn't kind enough to break down the numbers by city or ethnicity (which may be lacking in the poll itself, as they obtained it), but based on what it does report, the pollsters didn't talk to a single Kurd, who comprise about 15 percent of Iraq's population.

According to the AP, 55 percent of Iraqis reported to the pollsters they would feel safer if U.S. troops immediately left. This is said to indicate a big tumble in Iraqi affection for their liberators, and the AP links it to the Abu Ghraib scandal.

But for comparison, look at the USA Today/CNN/Gallup nationwide poll of 3,500 Iraqis, taken in late March and April, before the scandal broke big.

In answer to that poll's question, if the Coalition left today, would you feel more safe or less safe, only 40 percent said "more safe" or "no difference." But in the Kurdish areas, only 5 percent said "more safe." In Baghdad, meanwhile, at that time 47 percent said "more safe" or "no difference." When you factor out the Kurds, the gap between the old poll and the new one closes considerably.

Bear in mind, too, that the earlier poll was not limited to cities, and the later one was. The direct experience of violence involving Coalition troops likely is greater in cities than in the countryside, whereas the degree of rebuilding and improvement is probably measurably greater, at this point, in rural areas than in urban ones.

And of course, it's always possible that the Iraqis now say they would feel "more safe" without the Coalition around because they have more confidence in the local police and military, which would mean the Coalition is doing its job. The AP article even mentions that this confidence seems to be apparent in the new poll. But it keeps that news far from its reporting on the "more safe or less safe" question. That would seem too much like "good news."

According to the new poll, "Ninety-two per cent of the Iraqis said they considered coalition troops occupiers." In the March/April Gallup poll, that figure was only 71 percent. However, in answer to the "occupiers or liberators" question, back in March and April 82 percent of Baghdadis said occupiers; in Shi'ite areas and Sunni areas it was 80 percent. (In Kurdish areas, it was only 1 percent.) Once again, the gap is less than it looks, if you equalize the samplings in the two polls.

This question is a false dichotomy anyhow; the U.S. and coalition troops certainly can be seen as both occupiers and liberators. The Gallup poll was nuanced enough to phrase the question to allow for this: "Do you think now of Coalition forces mostly as occupiers or mostly as liberators?" Without knowing the exact phrasing of the new poll's question, it's difficult to compare, again, but the AP article seems to indicate it was offered as a straight "either/or" proposition.

Without knowing which percent of the latest poll came from which cities (Basra, Diwaniyah, Hillah are Shi'ite, Baquba is Sunni Triangle and Mosul is mostly Sunni), it's hard to make a direct comparison, but clearly if you lop off that overwhelming positive response from the Kurds and rachet up the total negative answers from the March-April poll, the change is not as great as it first appears.

Here's another bit of warped AP reporting:

Anger at Americans was evident in other aspects of the poll, including a rapid rise in popularity for al-Sadr, the Muslim cleric who has been leading insurgents fighting U.S.-led coalition forces.

The poll shows more people supporting Sadr. This much we agree on. But how does the AP get off calling that "evidence" of "anger at Americans"? Since half the cities are in Shi'ite areas, it could as easily, it seems to me, indicate people looking over their shoulders at Sadr's gunmen walking through their neighborhood, and giving the "right" answer.