USS Clueless              Voyages of a restless mind

Main: normal long no graphics Contact Log archives Best log entries Other articles Site Search

Stardate 20011220.1259 (On Screen): Sometimes the results of a survey depend very heavily on how the questions were asked, and who was asked. The International Herald Tribune has done a survey of Americans and others around the world asking questions about the September attack and about the war which resulted. As should not be too surprising, the Americans they talked to had different opinions that did the people from around the world. Of course, as soon as you see the description of who they talked to, alarm bells go off: "opinion leaders". How do they decide who an "opinion leader" is? (And just how successful have "opinion leaders" actually been at leading opinion? The evidence recently is "not very.")

Asked if many or most ordinary people consider U.S. policies to be "a major cause" of the Sept. 11 attacks, fewer than 1 in 5 respondents from America said they do. But in the rest of the world, nearly 3 out of 5 agreed that they would.

I'd love to see how that question was actually phrased, and I'd love to see the answers they allowed, because I think that given a choice about 98% of Americans would have answered this question "I don't give a damn." But that probably wasn't a choice. Here's the critical question on this subject I bet they didn't ask: Does this convince you that those American policies were wrong? I suspect that the vast majority of Americans would have answered "no" to that. The fact that someone else objects to our policies, perhaps even violently, doesn't in itself prove that those policies were wrong. American foreign policy does not have as its goal making us popular in the world.

It found strong support for the U.S. war on terrorism, when the fight was described in broad terms. About 6 in 10 of non-Americans said that most or many ordinary people believed that "the U.S. is doing the right thing for the world by fighting terrorism." Support rose to 9 in 10 in Western Europe.

But support tumbled when respondents were asked whether the United States and its allies should attack countries like Iraq and Somalia if they are found to have supported terrorism. While half of American respondents said those countries should be attacked in that case, the comparable figure was less than 3 in 10 outside the U.S.

That's hardly surprising; there are a lot of people out there who are suddenly feeling distinctly vulnerable. The US was much less scary when it was a "hyperpower" which sat back fat, contented and happy. Now the US is thoroughly aroused and beginning to use its military and economic might actively, and its military might is revealed as being even more formidable than many people realized. The US committed only a quarter of its carrier battle groups, perhaps a tenth of its air force, and less than half a division of ground forces and annihilated the Taliban. What nation could stand if we really exerted ourselves? So of course they're worried.

A lot of people out there are hoping against hope that the US will, once having pummeled the Taliban, again return to complacency. That isn't going to happen; American voters think that would be the height of idiocy -- and we're the only ones whose opinions count in making those decisions. (Sorry, "world opinion leaders"; that's the breaks.) So many of those people are worried about what else we may do before we're finished, in their nations or in nations that they have economic or political interests. They really wish we wouldn't. Tough shit.

Among Americans, 7 in 10 believed that the United States is taking into account its partners' interests in the fight against terrorism. But among those surveyed abroad more than 6 in 10 said instead that the United States was "acting mainly on its own interests."

The truth is somewhere in between, but mostly in line with the "mainly in its own interests" side. In some regard there has been extensive consultation and cooperation with allies, particularly in sharing of intelligence and in work to find and take out cells and agents of al Qaeda and to seize bank accounts and to shut down fund raising operations. In military matters, the US has been keeping its allies very loosely informed but has not been asking advice nor waiting for permission, and has been setting the goals for the military operations largely without consultation. The goals are not being deliberately chosen to screw over third parties, but their interests are secondary.

I'm extremely skeptical about this survey because I don't believe the sample. Merely by the fact that they claimed that they talked to "opinion leaders" that suggests that this was not a random sample. They specifically chose the people that they talked to. So who picked the sample, and what criterion did they use, and was it biased? Of course it was biased; the question is how. In other words, were "opinion leaders" people that the pollsters wished were leading the opinions of the world, those whose opinions coincided with the pollsters themselves?

They only spoke to 275 people. If that were the number consulted in a single nation, that would be a little light (typically these kinds of polls try to reach about 1200). But this is 275 people in something like 20 nations, and they're sometimes breaking the result down by region. As a result, when they say that 6 out of 10 respondents in Islamic countries considered the US attack to be an overreaction (which they did) that may be based on as few as 30 people (who were not chosen randomly), which makes the result completely meaningless.

Whenever I see a poll which clearly serves a particular political position and which is severely methodologically flawed, I immediately have to wonder whether those who commissioned the poll had an ulterior motive, and in this case I think there was one. The purpose of this one is obvious: it's yet another attempt by European "opinion makers" to rein in the US and convince us to return to safe-and-sane multilateralism. (discuss)

Update: Iain Murray writes to point to this article which gives more details about this survey. It's even worse than it looked on first examination; the sample is ridiculously small and not statistically significant. This survey cannot be extrapolated in any meaningful sense. The US result was based on 40 people, and in most nations they only talked to 10 folks. And the ones they talked to were not even remotely randomly chosen. Another problem with it is that it was done over a period of a month extending from November 12 through December 13, and during that interval the situation was very fluid; it's not clear that they're even getting consistent results.

Another thing is that the original article misreported a lot of what was said. The respondents were not asked whether they thought that US policies were partially responsible for the attack. They were asked whether that was the consensus among people in their nation -- irrespective of their own personal opinions on the subject.

It's been a long time since I've seen a survey which was so badly designed.