'The core sample group is a representative 10,000 people from England and Wales. But there's also an ethnic minority boost sample of 5,000 and a Muslim boost sample of 1,200 "to ensure the views of these groups are robustly represented".'Hold on this is the latest Citizenship Survey from the Department for Communities and Local Government for April-September last year and the sampling is skewed! Why is the sampling so skewed and what effect does this have on the results? The headline finding according to The Spectator is that 'about 85 percent of people think that their neighbourhood is cohesive, community-speak for the absence of overt ethnic and religious tension.' but as Melanie McDonagh writes
'presumably, there's an overlap between ethnic minorities and Muslims. That means, though, that the survey gives disproportionate weight to the views of minorities and Muslims as opposed to the rest of the population. '
What on earth is going on at the heart of government in this country that official statistics are skewed 'to ensure the views of (certain) groups are robustly represented'? I find this discovery rather shocking, do you?