Religion is the single most powerful predictor (among eight factors) of an individual’s likelihood to say that they would never vote Conservative, with 44% of those with no faith rejecting the possibility of supporting the Tories, compared with 35% of Muslims, 32% of Christians, 26% of Sikhs, and 19% of Hindus. Just 15% overall stated that they would never vote Labour, including 7% of Muslims, 6% of Hindus, and 5% of Sikhs.
These findings emerged from a new report – Degrees of Separation: Ethnic Minority Voters and the Conservative Party – released by Lord Ashcroft (businessman, author and philanthropist, and ex-Deputy Chairman of the Conservative Party) on his blog on 28 April 2012, and featured in the Sunday Telegraph the following day. The 50-page report and 159 pages of data tables are available at:
10,268 adults living in the census-defined Middle Layer Super Output Areas with the highest concentration of black and minority ethnic (BME) persons were interviewed by telephone on Ashcroft’s behalf between 24 October and 4 December 2011. These areas, which are largely urban and working class in profile, account for 70% of the BME population of England and Wales.
The sample included 4,590 self-identifying Christians, 513 Hindus, 1,747 Muslims, 308 Sikhs, 761 other religions, and 2,124 of no religion. Christians and those of no religion were overwhelmingly white (77% and 84%), with 94% of Sikhs and 90% of Hindus Asians. 67% of Muslims were Asian, 12% white, and 11% black.
In the May 2010 general election 24% of Christians recalled that they had voted Conservative, against 18% of Hindus, 15% of Sikhs, 13% of no religion, and 12% of Muslims. The Labour vote had been highest among Muslims (41%) and lowest for Christians (28%), with Hindus and Muslims on 35% and 37% respectively. Non-voters were above the average of 29% for Muslims (31%) and those of no religion (33%).
In the event of a general election being held ‘tomorrow’, only 6% of Muslims said they would vote Conservative, 7% of Sikhs, 13% of no religion, 14% of Hindus, and 20% of Christians and other religions. 51% of Muslims, 48% of Sikhs, 37% of Hindus, 33% of Christians and those of no religion, and 31% of other religions inclined to support Labour.
Asked whether they identified with one political party as consistently representing people like themselves, only 5% of Muslims and 9% of Hindus and Sikhs replied the Conservative Party (against 15% of Christians and 8% of no religion). The affinity of 47% of Muslims, 46% of Sikhs, 38% of Hindus, 30% of Christians, and 23% of those with no religion was with the Labour Party.
Of all religions, Muslims were especially prone to say that the Conservative Party (41%) and David Cameron (38%) do not really care about people from different ethnic or religious backgrounds. However, these figures were somewhat exceeded by those professing no religion (44% and 39%). 8% of Muslims saw the Conservatives and Cameron as actively hostile to individuals from different ethnic or religious backgrounds.
Correlated component regression analysis was used to isolate specific views most often associated with an unwillingness to vote Conservative among BME religious groups. For Muslims the strongest drivers were found to be a perception that the Conservative Party does not stand for fairness, is actively hostile to people from different ethnic and religious backgrounds, and that its policies have shown this to be the case.
Non-Christians and those without faith were somewhat more likely than Christians to believe that the economy and employment were the most important issues facing the country. Muslims (57%) were far more likely to prioritize education than Christians (38%), and, together with those of no religion, much less likely than Christians, Hindus and Sikhs to see control of immigration as a key topic (one-quarter for the two former versus two-fifths for the three latter groups).
It must be reiterated that the main purpose of Ashcroft’s survey was to probe BME attitudes to the Conservative Party. Its underlying sample was not nationally representative, especially in terms of social class and locality. Also, for BMEs the Ethnic Minority British Election Study (EMBES) is an important alternative dataset. Some preliminary analysis of this by religion was posted on BRIN’s website last year by Ben Clements and Maria Sobolewska at: