Posted by 5cc
August 31st, 2009
Usually, at this time of year, the Mail is busy writing up stories on the back of newly released immigration figures. Last year, we were treated to stories about how many white people were leaving the country, the year before we had big spreads about the number of UK citizens leaving while immigration figures were up. This year, though, the immigration figures were largely positive from a Mail point of view. More foreigners leaving, fewer arriving, fewer UK citizens emigrating and so on. The paper had to focus on the number of children born to mothers who were from overseas to frighten us with instead.
So today’s ‘One out of every five killers is an immigrant‘ looks a little out of place. That may be because it takes three weeks to get a reply from the police to an FOI request and the hack who wrote it was anticipating rather different immigration figures to be published when he made the request.
Whatever the reason for the story, it’s an example of a very common and very misleading tactic that the Mail (along with the Express) engages in when it wants to make us frightened of foreign criminals. You can see the same tactic used in ‘One in six rapes committed by foreign attackers, shock police figures reveal‘ from April this year (although that story was churned directly from the Daily Express) and ‘Foreigners carry out one in every five killings in Britain, police figures reveal‘ from April 2008. You’ll notice that the last article there reveals that the ‘one in five’ figure isn’t actually news, since it was reported over a year ago.
Here’s how the tactic works.
First, the paper contacts every police force in England & Wales and asks for stats showing how many of one crime or another has been committed by foreigners. Then the paper then calculates how this translates into percentages across the UK.
Here’s why the tactic is misleading.
1. Police forces don’t have completely reliable figures for how many foreigners commit crime. All they have is a box for ‘nationality’ on arrest forms, which are voluntary and never checked.
2. Not every police force responds, but the Metropolitan Police always does. The Metropolitan Police arrests more people – and more people who enter something other than British into the ‘nationality’ box on their arrest form than any other force. The current Mail article talks about there being 371 individuals accused of murder or manslaughter last year, with 233 of them being in the Metropolitan Police area. This will completely skew the numbers for the rest of the country, even if they’re proportionate for the London area.
3. The paper does not compare the number of arrests of people who enter something other than British into the ‘nationality’ box on their arrest from in the responses they get to the number of people born overseas in the areas they have replies from. Instead, they compare it with the whole country.
To illustrate this with an extreme hypothetical example – London has an immigrant population of around 30% and arrests more people for murder or manslaughter than any other police force. Let’s say that in one year, the number of homicides in London that people who enter a non-British nationality in their arrest form are completely proportionate to the number of people born overseas in the area – 30%. In that same year, there are no homicides anywhere else in England & Wales. We now have a scary ‘Foreigners commit a third of killings in the UK but only make up 10% of the poplulation’ story. The trouble is – that’s completely proportionate in the actual area those killings took place.
Now, the paper does state that “In London, almost 40 per cent of those in such cases in the past year were from overseas, or of unknown origin,” which would be disproportionate were it not for the fact that the paper has decided to add everyone who didn’t enter anything into the ‘nationality’ box. As the article later reveals, including these people makes the total in London higher than the actual total across the country, which is impossible. Could it be that the hack has included this figure rather than the actual figure because the real one would make it too obvious that this article is misleading?
This ‘get an FOI request from police forces’ tactic will always return a scary looking overall average. Great for frightening the readers with – not so great for actually giving an accurate idea of how many crimes were committed by people from overseas.