When I first read of this poll (via a number of blogs) I was pretty incensed :
Israel has been described as the top threat to world peace, ahead of North Korea, Afghanistan and Iran, by an unpublished European Commission poll of 7,500 Europeans, sparking an international row.
The survey, conducted in October, of 500 people from each of the EU’s member nations included a list of 15 countries with the question, ‘tell me if in your opinion it presents or not a threat to peace in the world’. Israel was reportedly picked by 59 per cent of those interviewed.
The full text of the survey-results can be downloaded from here as a PDF file (3677kb).
Once I realized how flawed the poll is I calmed down again; cold-calling people on the phone, reading them out a list of countries and asking which of those is a threat to “world peace” (without even defining the term) isn’t good polling technique. Now, if the pollsters had visited the respondents at home and asked them which countries are a threat to world peace without naming any first the responses would have been more significant. But giving them a list to start with gives the questions a leading quality. Also, if you call people and ask them for their opinion on the fly you won’t get their well-considered opinion, since on this short notice they’ll mostly think of what they heard or read about the other day and respond according to that. So any country that is frequently mentioned in context of some conflict or war, as America or Israel is, is more likely to be named as a threat to peace as countries which aren’t as often in the news. And as it happens neither North Korea nor Iran are all that often mentioned in the context of war (more like oppression and misery), so why should anyone who isn’t attentively following the news call them threats rather than basket-cases?
The poll also suffers from an imbalance: The sample was 500 people in each country, so that Luxembourg with its 440.000 inhabitants had as much weight as Germany with 80 million inhabitants, which further distorts the results and devalues the respondents’ answers to the point of meaninglessness.
As another example of a meaningless poll you could take the one where about 80 percent of American respondents answered that Saddam Hussein was behind 911, even though nobody had ever officially made that claim. As with the European poll, people had simply thought of the speculations they had read or heard about and responded accordingly.
You could also look at this poll:
SHOCKING POLL: A MAJORITY OF AMERICANS CANNOT NAME A SINGLE DEPARTMENT IN THE PRESIDENT’S CABINET.
Washington, DC – Most Americans are unable to identify even a single department in the United States Cabinet, according to a recent national poll of 800 adults. Specifically, the survey found that a majority (58%) could not provide any department names whatsoever; 41% could. Only 4% of those surveyed specified at least five of the 19 executive-level departments, a figure comparable to the poll’s overall margin of error (+/-3.5%).
This poll doesn’t mean that Americans are ignorant, of course. Europeans wouldn’t have done any better if asked about their governments but it shows clearly how little substance there is to answers people have to provide on short notice.
I rest my case. :)