Spinning Statistics on Muslims

The PA headline reproduced in The Scotsman announces shocking news:

British Muslims ‘Back Terror Strikes on US’

Apparently, a poll has discovered that the Muslims of Britain support al-Qaida! However, if you decide to read on, you find that

Attacks on the US by al Qaida or other groups were viewed as justified by 13% of the 500 British Muslims questioned. Another 15% said they did not know whether such attacks are wrong or right.

In other words, as The Guardian (the newspaper that commissioned the poll) puts it rather less sensationally:

The ICM survey…shows that the overwhelming majority of British Muslims – 73% – are strongly opposed to terrorist attacks by al-Qaida and other organisations. But a small minority – 13% of British Muslims – disturbingly say they believe further such attacks on the US would be justified.

This interpretation is rather at odds with the impression given by the PA headline: a headline that many people will not look beyond, and which is currently on the religious news websites.

The Guardian also explains the methodology of the poll in more detail than the PA report gives us:

ICM interviewed a random sample of 500 Muslims between March 3 and 11 2004. The interviews were conducted by telephone throughout the country. ICM also interviewed a random sample of 1,014 adults aged 18 and over by telephone between March 10-11, 2004. Interviews were conducted across the country and the results have been weighted to the profile of all adults.

That’s fine as far at it goes, but among British Muslims there are Arabs, Pakistanis, West Africans, Turks, white converts and many other ethnic groups, as well as confessional differences between and amongst Shias, Sunnis and nominal Muslims. A random sample of 500 Muslims that may or may not represent all these strands seems to me of limited value.