I've just had a very interesting evening. At least, the results are interesting, not the 6 hours of manual data entry and checking that led up to them. What have I been doing?
Well, over the weekend I (and everyone else) wrote long pieces about the ICM poll of Moslem attitudes. Particular focus was paid by us all to the data about the possible introduction of Sharia law for various purposes. Those that expressed a preference were split evenly down the middle.
A lot of the commentary about this focused on the attitudes of the women interviewed, and quite reasonably assumed that they would have different attitudes to the men, and would be less likely to back Sharia law.
Today, ICM published the
full results on their web site. Lots and lots of tables of data, which actually allow us to answer some of the questions posed: about how attitudes change according to gender, age, social class, voting intention and location. And it turns out that that analysis shows up some interesting stuff.
The AnalysisA note about the analysis that I've done. Nothing complicated, of course. What I have done, though, is to build some "net" statistics: the
difference between those in favour and those against a proposition. It's the equivalent of the statistics produced on politicians on a regular basis.
So, if 21% of the sample group of 500 people think that the moon is made of green cheese, while 20% disagree, the net result is 1% in favour.
This approach has the benefit of ease of analysis. The sample is split according to a number of groups: men vs women, age, class, voting intention and location. Obviously these are separate analyses: the women are included in each other group, as are the Tories, the people from the Midlands and so on.
Helpful suggestions about additional approaches that might be useful always appreciated, because with this amount of data there may be some more posts on this topic coming out.
On to the results. I'll take the issues one at a time...
Sharia LawAs discussed above around
40% of the total sample were in favour, with an equal number against. Look at it on the segmental basis, though, and things change.
Women, for example, turn out to be
net in support of the introduction of Sharia law in certain circumstances (a net of 3% in favour) while
men are mildly opposed (a net of -6%). Those
under 35 are in
favour (+5%), while those
35 and over are opposed (-9%).
ABC1s are strongly opposed (-12%), while
C2DEs are quite strongly in favour (10%). Voting intentions are interesting too, with
Tories (-33%) and LibDems (-21%) very strongly opposed, but with
Labour supporters actually neutral on the issue (a net of 0%, believe it or not). There is quite strong support in the Midlands (+8%), with opposition in the South (-7%)
RadicalisationNext we look at whether Moslems have become more radical over the last year. Remember, here the
total figure showed
46% thinking that they had (a net of +13% over those disagreeing). But again, there is lots of detail.
Women (+20%) feel this much more than men (+7). The young feel it more than the old (+21% vs 4%). Class is less split (+18% for ABC1s, and 10% for C2DEs).
Voting intentions don't change things much (nets of +16% for Tories, +13% Lab and 23% LibDems). Interestingly, the Midlands
supports this idea much less (+3% only, vs +16% for the North and +17% for the South).
Are Women Covering Up More?Here we see some interesting differences too. The "No Difference" responses are the majority across the board: so let's remember that
for most it hasn't changed. However, where there have been changes perceived,
men feel significantly more strongly that women in their family are more often covered up in public (+24%) than women do (+14%). Are Moslem women telling their families "Yes dear", leaving the house covered, and then ripping off the veil when they turn the corner, one wonders? No real difference by age, here (+18% and +19%).
Class does matter: a net of +25% for ABC1s compares to a 13% for C2DEs. Tories have seen the least changes (+6%, compared to +16% for Lab and +17% for LibDems), and there has been much less difference in the North (+7%) than the Midlands (+22%) or the South (+25%).
Are Relations Getting Better?Mostly people responded to this one that relations between Moslems and others were about
the same: around the 50% mark for each group.
Women were markedly more pessimistic, though (-10% vs -2% for men), as were
those over 35 (-14% vs -8% for the younger group). Identical net figures for the two class groups, but Tories
(-27%) much more strongly negative than Labour (-5%) or LibDems (-17%). Oddly enough, while the North and South are both negative (-16% and -12%), the Midlands are actually positive (+1%), suggesting that relations there have actually improved.
I'll stop there, for now. I've got some more to do to this data set, and some pretty charts too, if I can get Blogger to co-operate. Tomorrow evening, if I can rouse up the courage.
What does this initial analysis tell me? A few things, I think.
First, we can't make assumptions about the thoughts of these groups. The women interviewed certainly don't appear to be taking a much softer view (although their answers about violence tend to be marginally less belligerent than the mens'). There certainly doesn't seem to be a major attack on Sharia from them. And let's note that this was done by phone, which is probably the most likely to get an unbiased response from this group (any fear of pressure from husbands is minimised by the fact that the husband can't hear the questions).
Second, there seem to be significant differences in perception between political groups (although there are many more Labour supporters in total). This is good news for Tories: support we're getting from this community appears to be based on a group with distinctive views, which may make it less under threat. However, some of the attitudes of the Labour and LibDem voters could raise some concerns in those parties...
And third, though we may find it difficult to understand, there is strong and genuine offense across the board caused by these cartoons: the only group where this figure drops below 80% of the respondents being offended is the Tories, and there it's in the high 70's. This isn't an argument for changing our views about the cartoons, by the way, but more one for better explanation of why free speech is in all of our interests. But that's for another day.