Careful how you read this BBC News report. Whoever wrote the title obviously didn't read the article.
In reality it should probably be
Few white friends for ethnic minorities. According to the article,
more than 90% of whites have
no or few friends who aren't white.
I'm going to pick an arbitrary definition of
few - let's say it means
three or under; as another ballpark figure, let's assume about thirty friends each in total (seems ever so slightly exaggerative, but what the heck).
This would mean that, for those people who are not in the 90%, i.e. those who don't have
few or no non-white friends, more than 10% of their friends are non-white. If a person has fewer than 30 friends (I can't think of 30 people I'd count as
few (up to 3) of them is an even larger proportion.
One of the findings of the 2001 Census was that, according to the BBC, 9% of Angles and Welshmen aren't white. Which means it's perfectly reasonable to have
few - i.e. less than 10% (assuming my guesstimations are about right) - non-white friends.
It's even more reasonable if you're only counting blacks - they make up 2.2% of Angles and Welshmen. So if I have one black friend, and fewer than 45 friends in total, I have a disproportionately large number of black friends.
Disclaimer: the above maths is probably invalid and rubbish.
The more interesting part is that
nearly half (let's call it half anyway) of non-white folk
say most of their friends are white. Which means 50% of non-whites don't say most of their friends are white. Once again, 90% of people in England and Wales are white. That's most. Statistically, everyone should be saying that most of their friends are white.
Of course, this shows that friendships aren't distributed the same way as the population. It seems that white people have a statistically probable number of non-white friends, while non-whites have an improbably high number of non-white friends. My theory would be that most immigrants would tend to socialise with other immigrants, and non-whites make up a larger proportion of immigrants than of the general population.
The CRE's chairman, Trevor Phillips, said he had been surprised by the extent to which the majority community still did not really know minority communities.
There isn't any real majority
community - just the society of which minority communities are part. Maybe people of those communities generally have more friends (as they're part of a community) hence the disproportionately high non-white friend count among non-whites.
Another suggestion might be that those communities are insular, but that's not borne out by the (alleged) fact that whites have a proportionate number of non-white friends.
The definite (not a trace of doubt) conclusion which can be drawn from this, then, is that white people in England and Wales aren't as friendly as everyone else.
So that headline should probably read either
Few black friends for whites; few black people around, or
Few friends for whites.
If only we had more minorities.