In a year that has already seen three deadly terror attacks on UK soil in three months, the spread of Islamist extremism is at the forefront of many people's minds.
In recent years, polls and studies have sought to discover how Britain's Muslims feel about extremism and the terrorist threat.
Why the polls never give a full picture
The Week
Escape your echo chamber. Get the facts behind the news, plus analysis from multiple perspectives.
SUBSCRIBE & SAVE
Sign up for The Week's Free Newsletters
Join more than 350,000 subscribers and keep yourself informed with a selection of The Week’s most interesting, enlightening and entertaining stories - plus daily puzzles.
Join more than 350,000 subscribers and keep yourself informed with a selection of The Week’s most interesting, enlightening and entertaining stories - plus daily puzzles.
Latest Videos FromThe Week
As with many questions that rely on polling data, the answer depends on what, who and where you ask.
Asking British Muslims their opinions might seem like a simple way to get a picture of how the community as a whole feels towards extremism, but polling is far from straightforward.
One of the most recent large-scale surveys of social attitudes within British Muslim communities, carried out for Channel 4 by ICM in 2016, found one per cent "completely" sympathised with people who took part in suicide bombings to "fight injustice", while three per cent had "sympathy to some extent".
Meanwhile, 0.5 per cent said they fully sympathised with those who committed terrorist acts to further a political cause, while another 3.5 per cent had "some sympathy".
However, ICM's methodology came under fire from the Muslim Council of Britain, who said the poll's sampling meant it was not representative of the British Muslim community as a whole.
Writing in The Guardian, Miqdaad Versi, the council's chief, said that by choosing to question people only in areas where at least a fifth of the local population was Muslim, the survey was skewed toward "poor and religiously conservative" Pakistani and Bangladeshi enclaves.
Understanding the data
It is impossible to draw meaningful conclusions from surveys of British Muslims without also examining the opinions of non-Muslims.
While the 2016 ICM poll ran a control group consisting mostly of Christians and non-religious people, no media outlet actually compared that data.
If they had, they would have found that while four per cent of Muslims had some degree of sympathy for suicide bombers and terrorists, one per cent of the control group also had "some sympathy".
While 13 per cent of Muslim respondents said they could understand how a British Muslim could become attracted by radicalism, 26 per cent of the control group said they could.
Open to (mis)interpretation
Once the data is out of the statisticians' hands, it is subject to misinterpretation and misrepresentation in the press.
In 2006, an ICM opinion poll was widely reported as showing that 20 per cent of British Muslims sympathised with the bombers who carried out the 7/7 attack in London.
However, the actual phrasing of the question asked if respondents sympathised with the "feelings and motives" of the suicide bombers. Asked if the bombing itself was justified, 99 per cent said no.
More recently, in 2015, The Sun claimed one in five British Muslims sympathised with jihadis.
Its source was a poll by Survation which specifically asked if participants had any sympathy for "young Muslims who leave the UK to join fighters in Syria".
Five per cent said they had "a lot" , while another 15 per cent had "some" sympathy.
But their compassion "lay with the young Brits, not the Isis jihadis", says The Spectator.
Crucially, one in six non-Muslims asked the same question also expressed sympathy.
Why should "one person's world view [be] automatically exempt from scrutiny", asked the magazine, while a Muslim's is viewed with suspicion "just because part of it involves going to a mosque?"
Press regulator Ipsos ruled the Sun's headline was misleading and the paper was forced to publish a statement acknowledging that.
