Does more diversity really lead to unhappiness?
9:07 am - June 25th 2009
Tweet | Share on Tumblr |
Excitingly for data-mining weirdos, the Department for Communities & Local Government has released data on various happiness-related statistics broken down by local authority. What else is broken down by local authority, that I’ve written about recently? – yup, ethnicity statistics.
One of the questions asked in the poll is “% who agree that their local area is a place where people from different backgrounds get on well together”. If the theory that BNP voters are driven by fear of gangs of steaming Somalis, murderous mullahs and crack-dealing Caribbeans were correct, then you might expect there to be some kind of negative correlation between Nick Griffin’s multi-ethnic nightmares and the belief that ’round here things are pretty much OK’.
So I dug up my trusty ethnicity-by-local-authority dataset, and plotted a graph of minority population (meaning ‘not white British’) versus happy-with-race-relations-round-here:
Pleasingly, r^2 here is zero, which means there is absolutely no correlation at all, not even a tiny bit, just none whatsoever between quality of race relations and proportion of immigrants.
The rest of the data’s here, if you want to download it – the general trend is a slight (r^2~ 0.1) negative correlation between minority population and most happiness statistics. At some point I’ll control by income and deprivation: but based on the work I’ve already done on this, the uncorrected correlation is so small that I’d be amazed if minority population had any substantial impact on any of the above.
Returning to the BNP vote question, the data also shows that Barking & Dagenham and Newham are the most miserable places in the country, based on “% who are satisfied with their local area as a place to live”. Newham has a 66% minority population; Barking has a 19% minority population. About 15% of Newham’s white British voters voted BNP [*], as did about 15% of Barking’s white British voters.
This is reasonable evidence for the proposition ‘people vote BNP if their life is extremely miserable, irrespective of whether or not this has anything to do with immigrants’ – which strongly implies the best way to combat the BNP is to help make people’s lives less miserable rather than to kick out immigrants…
[*] I’m assuming that ‘people who aren’t white British won’t vote BNP’ is a reasonable generalisation.
Tweet | Share on Tumblr |
John Band is a journalist, editor and market analyst, depending on who's asking and how much they're paying. He's also been a content director at a publishing company and a strategy consultant. He is a regular contributor to Liberal Conspiracy and also blogs at Banditry.
· Other posts by John B
Story Filed Under: Blog ,Equality ,Race relations
Sorry, the comment form is closed at this time.
Reader comments
Wouldn’t it be great if the main stream media showed correlation coefficients and then offered a file containing the raw data.
John,
In the lates happiness survey – it seems like most of the countries mentioned are very homogenous eg Denmark, Switzerland, Austria, Iceland, Finland, Bahamas, Bhutan, Canada,Luxenburg, Sweden.
Hi John,
This is interesting.
Looking at the data from Ealing Council’s 2009 residents’ survey, there is no correlation between satisfaction with local area and what they call “ethnic fractionalisation”, which reinforces your point. There is correlation between satisfaction and levels of deprivation (the less deprivation, the greater satisfaction), which again is no great surprise but suggests that the way to increase happiness is to reduce deprivation, not reduce ethnic diversity.
I concur with the above poster who said wouldn’t it be great if mainstream media posted this?? Or even political parties if they genuinely claim that they ‘care’ about the issues we have with racism…
Surely this should be bang on Labour’s website! This should be what they shoul’ve been discussing while our canvassing or delivering!
But I somehow think it is not the media nor political parties interest to not create fear and unrest, as it seems that is the only way they can maintain empirical control over the oi polloi..
‘Wouldn’t it be great if the main stream media showed correlation coefficients and then offered a file containing the raw data.’
It would be an interesting exercise to take the graph and raw data and turn it into a mid-range newspaper article. On the graph, you can see exactly which data points they would pick to talk about:
—
In bottom_right_1, … Local housewife Jane Doe, 48 … scared .. gangs .. knives’.
At the other end of the country, in bottom_right_2, the picture is the same …
Not every region has these kind of problems. In top_left_1, ‘…
—
A data set that offers no correlation is hardly even a challenge to cherry-pick a narrative from. You can convince most people with 3 anecdotes, and sceptics with 5.
Trained scientists/engineers might take 7.
Great post John.
Sadly it is easier for government to restrict immigration than take steps to reduce deprivation. They can say they are “doing something”, they are encouraged by the press, and they are motivated by public perception rather than what will actually improve people’s lives.
I thought it was quite well established that higher diversity led to lower levels of trust, which isn’t exactly the same thing as hapiness but has implications for social cohesion.
More to the point, surely it’s up to the advocates of diversity to show that it “enriches” people’s lives, rather than for it’s opponents to demonstrate the opposite.
I find this sort of research inherently suspect. But it is worth noting there is a difference between being happy with race relations (which is pretty easy to attain so long as there isn’t too much competition over resources) and Robert Putnam’s metric of “social capital”, which diminishes significantly in more heterogenous populations.
Of course, I don’t trust this concept enough to make it worth seriously persuing policies that increase “social capital” (rather than more general prosperity), and don’t believe we should be steering towards a reactionary immigration policy on that basis. I just find it ironic that a concept that was pretty much invented by sociologists to bash the free market turns out to endorse homogenous communities surprisingly consistently.
@1 and followups, if we’re going to be fair, the Grauniad did post the happiness data – they didn’t post the ethnicity data, but since that was my idea it’d be harsh to blame them for it.
@2 Luxembourg is 37% ethnic minority, Sweden 14%, and Denmark 10%. Canada is 20% ‘visible ethnic minority’ (ie not white) and has the world’s highest immigration rate. So I respectfully suggest you’re talking absolute nonsense.
@7 it’s a common talking point. Whether it’s ‘quite well established’ is another question altogether, and one that doesn’t seem to be true on the basis of this data.
Interesting stuff. It looks to me a bit like people from different backgrounds start getting on slightly worse when the ethnic minority population hits about 10%, and then matters sort themselves out once it reaches a critical mass above about 20%, though without knowing what those three outliers are, it’s hard to say. I guess the data set isn’t really big enough to do the average for each percentile range.
Surely there are too many externalities for this to be a meaningful graph, though – areas where people from different backgrounds don’t get on well together might have lower growth in the ethnic minority population because, having that reputation, people from ethnic minorities are less likely to live there. Equally, horrible places where everyone is miserable already and therefore nobody gets on to start with because there are no jobs are presumably less likely to attract economic migrants.
http://www.crimereduction.homeoffice.gov.uk/activecommunities/activecommunities76.htm
A few years old but brings up the trust issue mentioned by Matt Munro above.
“The findings show that there is no significant relationship between diversity, levels of trust and community participation. However, it did show that people living in areas of high ethnic diversity had significantly lower levels of generalised trust (trust of others in their neighbourhood).
Similarly, the data showed that there is no statistically significant relationship between socio-economic diversity and all the measures of trust.”
the trouble is, matt, that it’s been proven time and again, moseso by these stats, that multiculturalism has no negative effect. We also know that immigration largely has no effect economically.
Why is it either our job to make a case for the benefits of multiculturalism or that multiculturalism should indeed “make a positive impact”?
Surely it’s the best thing for all if a diverse nation has no overall positive or negative effect on the nation at the least, and given this it must be your job now to point out why the hell something that isn’t a negative should be stopped at all.
“multiculturalism has no negative effect.”
I don’t think the above has anything to say about multiculturism either way, it has nothing to say about how how seperate or homogeneous these people are in terms of culture, just ethnicity. The standard criticism of multiculturalism, (by the sane rather than the “we just don’t like darkies” crowd), is that it encourages isolation and makes relations between ethnicities more fraught. Assuming, for the sake of argument, that you have similar levels of multi/uni culturalism across the board it might affect the hight of the line but not the gradient.
John B: Good article but I reserve the right to be a deeply suspicious SOB when it comes to “happiness stats”. Not that I’m alledging some sort of “Liberal Conspiricy”, I just think it’s a difficult thing to measure accurately.
@7:
surely it’s up to the advocates of diversity to show that it “enriches” people’s lives, rather than for it’s opponents to demonstrate the opposite.
Erm, no, not in a free society.
BNP voter: I’m unhappy about immigration.
John: no you’re not; let me show you this graph…
cjcjc: “I don’t like facts.”
@16
LOL
“@7:
surely it’s up to the advocates of diversity to show that it “enriches” people’s lives, rather than for it’s opponents to demonstrate the opposite.
Erm, no, not in a free society.”
Exactly – and if you want to change a free society you have to show how that change is beneficial, otherwise you are imposing the will of a minority on the majority…
“the trouble is, matt, that it’s been proven time and again, moseso by these stats, that multiculturalism has no negative effect.”
Nor does herpes, but that’s not an argument for spreading it.
“We also know that immigration largely has no effect economically”.
Then why do it ?
Slightly off-topic but interesting as Canada was mentioned above:
http://www.immigrationwatchcanada.org/index.php?module=pagemaster&PAGE_user_op=view_page&PAGE_id=1495
Matt Munro, are you the Matt Munro that is serving in Afghanistan and is mentioned in the Groan almost every day?
http://www.guardian.co.uk/world/2009/jun/24/afghanistan-black-watch-double-life
@19 – if it’s a free society, how can you stop it?
“Then why do it ?”
Why not do it?
More to the point, surely it’s up to the advocates of diversity to show that it “enriches” people’s lives, rather than for it’s opponents to demonstrate the opposite.
It may be an awful cliche, but: I like Vietnamese food, and I live in a city that, thanks to its ethnic diversity, has an awful lot of Vietnamese people; some of them are jolly good at cooking Vietnamese food; and some of _them_ then proceed to sell me it at reasonable prices. Thus my life is enriched.
But what if you didn’t like Vietnamese food, you liked pie and chips, but the chip shop closed down because the local populance (now multi-cultural) no longer supported it. By your own logic your life would be similarly un-enriched would it not ?
It’s a specious argument anyway – I like curry and you could get a decent curry in any UK city long before anyone could even spell multiculturalism. Something being avaiable and something being shoved down your throat are 2 very different propositions
“But what if you didn’t like Vietnamese food, you liked pie and chips, but the chip shop closed down because the local populance (now multi-cultural) no longer supported it.”
Boo hoo, cry. That’s freedom for you – sometimes stuff *you* don’t like happens. Besides, Tesco will have stepped in to fill the pie and chips market, so get over yourself.
Pretty much everywhere I’ve lived (north and south), the chip shops have been run by people of Chinese or Greek descent anyway…
@28 – And Turkish round my way.
Cue “But why should I have to buy my pie and chips from a muslim. It’s just not fair“.
Not to mention the fact that large multinationals do more to undermine British culinary traditions and farming than any wog ever did. What if I like my local farmers’ market but Tesco drives it out of business?
You’ll be waiting a long time for Matt Munro to object to that because it’s the “free market” and “choice”.
Usual right-wing fuckwit then, basically.
Also a total hypocrite.
“Boo hoo, cry. That’s freedom for you – sometimes stuff *you* don’t like happens. Besides, Tesco will have stepped in to fill the pie and chips market, so get over yourself.”
How is it “freedom” if something I want is no longer avaiable ? Last time I checked we lived in a democracy which means stuff I do/don’t like is as important as stuff migrants do/don’t like, except that when I deign to express that opinion it’s called racism by hypocritical middle class wankers like you.
@ So liking pie and chips make you a “right wing fuck wit” – go and drown in mung beans you twat – I’M NOT EVEN BRITISH
“@28 – And Turkish round my way.
Cue “But why should I have to buy my pie and chips from a muslim. It’s just not fair“”
So all Turks are Muslims ? And you’re calling me a racist ?
I hate to be a bit of a pedant about this but . . . No- bollocks to it- I LOVE being a massive geek about stuff like this: Your statistical test is a linear regression, not a correlation. A correlation will only test an association and not a functional relationship between variables. As you are testing a hypothetical model based on the two variables being functionally linked a regression is appropriate here.
However, your model assumes that Happiness With Race Relations [HWRR] is a function of the proportion of non White British in the region. This is clearly flawed as it assumes that all such people have a homogenous affect on the happiness of White British people. Living next door to an Irish family is very different to living next door to Finsbury Park Mosque (I struggled for a few minutes to come up with a non-Islamic example of potentially undesirable neighbours and, after discarding Greek students, Chinese cockle pickers, Somali asylum-seekers and Russian disco-fiends [don't ask] I decided it was a relevant one and went for it). Also, the actual question asked was “do yuo think that your local area is a place where people from different backgrounds get on well together?” “Backgrounds” can be interpreted by respondents to cover the entire diversity of personal history and therefore many respondents will not have interpreted this, as you have, as a strictly ethnicity-based enquiry.
That’s not to say I disagree with you about misery being a significant driver of BNP-voting. That, surely, is the very definition of a Protest Vote. Of course there are cunts like Griffin and that other despicable oik that got elected as an MEP who clearly vote BNP for ideological reasons but I hope they’re the minority. It would be interesting to ascertain what degree of fidelity BNP voters have to the party. Whether most of them only vote the once, at a low point in their lives and in protest, and then have an epiphany, recant and resolve never to be such a twat ever, ever again.
Word.
@Matt Munro: As I said before, get over yourself, cunt.
The Vietcong around here jusy grow weed they do sod all else just grow and sell weed.
Reactions: Twitter, blogs
-
Banditry » Blog Archive » Banditry elsewhere
[...] a new piece up at Liberal Conspiracy on happiness, tolerance and migration data. Which is fun if you like that sort of [...]
-
Patrick Hadfield
Liberal Conspiracy has a post on happiness and diversity – http://tinyurl.com/n2at8u – particularly useful for links to stats on happiness
[Original tweet]
Sorry, the comment form is closed at this time.