Sunday, 28 February 2010

The Daily Mail and the art of making the facts fit the editorial byline

The following post comes from the excellent Five Chinese Crackers, who have turned debunking the vitriol of the Daily Mail into an art form. With immigration certain to be a hot button issue at the general election, the continuing brutalisation of migrants in detention centres, and the Tories already looking to "unofficially" co-opt the BNP's clothing, such a dose of reality is long overdue for the tabloid media.

Regular readers will already know what the Daily Mail does when statistics are released that do not fit with what the paper wants to tell its readers about immigration, but James Slack (the Home Affairs Editor, no less. Ooh - get him) gives us a lovely lesson for new readers and old alike. Aren't we lucky!

Statistics were released on Thursday that should by rights please the Mail. You can find them here and here. A clue as to what they might be is in the page title of the article on the website (top left in the picture below the fold), which is strangely not the healdine of the story anymore:

See, we can't have that sort of headline. Goes against the narrative!

The new headline is 'Two passports a minute are given to foreigners as 1.5m issued since Labour elected'. It focuses on one of the few things in the latest statistics that the Mail can scare us with, since almost everything the paper wants to see falling fell and most of the things it wanted to see rise rose.

It's also not true. There are 1,440 minutes in a day, 525,600 in a year and 6,307,200 in 12 years. If two passports a minute were given to foreigners (interesting they're still referred to as 'foreigners' and not new British citizens) between 1997 and 2009, we'd be lookiing at a total of 12,624,400. That's twelve and a half million, not one and a half.
Here's what the article says:

Passports were given to foreigners at the rate of two a minute last year.
Officials approved a record 203,865 citizenship applications, 58 per cent more than in 2008.
Another 190,000 immigrants were given the right to settle in the UK in 2009 – a rise of 30 per cent on the year before.
Ah. So we're not talking about the time since 1997 like the headline implies - we're talking just one year.

Still, how many minutes were there in a year? 60 X 24 X 365 = 525,600. For two a minute, we'd be looking at 1,051,200. Five times the number we have. Unless we add the number of people given grants to settle, who I don't think are elegible for passports. Even if they are though, we're still looking at more than double the number we have to equal 2 a minute.

We have a Slack maths classic. Not quite equal to making 55 people seem like over 10,000, or adding together figures he must have known would give a misleading total before adding almost a billion extra pounds to the total - but it's a pretty good one.

Still, the numbers have obviously risen since last year. Why is that? The Mail says:

Officials claim the massive rises during the past year may have been caused by migrants rushing to beat the supposedly tougher system of earned citizenship due to start next year.
Okay. Maybe they do. Maybe they don't. Whaddayagunnado? But the actual statistics document, the one the figures presumably come from, says:

The number of decisions made in 2009 has recovered from the comparatively low level in 2008 when staff resources were temporarily transferred from decision-making to deal with administration of an increase in new applications.
Left out the bit that explained how the number of decisions made last year was 'relatively low'. Managed to keep the bit about the big increase in applications though. Funny that.

Notice also how he manages to say 'supposedly tougher' before telling us that the waiting period rises from 5 years to 7 or 8 years, which is most certainly tougher.

There's a scare quote from Damian Green before we move on to more statistics, which are:

A raft of statistics released yesterday showed that huge numbers of students continue to pour into the UK – despite concerns about bogus colleges and visas.
In the final three months of 2009, 61,715 student visas were issued – an astonishing rise of 92 per cent on the same period in 2008.
Alarm bells start to ring when you see the time frame shifted and zeroing into specific types of visa. They ring even louder when you check the stats and see there's a new category for student visa that wasn't there the year before.

There could be reasons for this rise that are explained by the introduction of the points based system. Maybe it covers more people than the old Student Visa System. Maybe it was rubbish at doing what it was supposed to do - make it tougher for people to cheat the system when they're not really a student. Maybe the new system created a backlog of applications that were dealt with later in the year. There seems to be an anomaly here. In Q2 there was a drop in the number of student visas granted of 10,000 compared to last year. Looking at the year as a whole:

The number of student visas issued to main applicants was 273,610, an increase of 31 per cent compared with 2008 (208,800).
So we have a much smaller rise than for the last quarter, which is of course why that was focused on instead of the whole year. Put this straight after the guff about passports and it makes it look like immigration's rising, right?

We're approaching the bottom of the article, so here's where we get the more positive (from a Daily Mail point of view) numbers. here's what we get:

The figures revealed a shift in the source of the arrivals.
Notice the wording there. Slack's about to start talking about the drop in the number of Eastern Europeans, and he segues between what he's already said by mentioning a shift in the source of arrivals.

He does this directly after talking about things that have risen (and exaggerated some of those by pretending they were at a rate of 2 a minute), which makes it look as though the number of 'arrivals' overall has risen, just from places other than Eastern Europe. In fact the number of people arriving in the country for more than a year has slightly fallen.

Welcome, ladies and gentlemen, to the tabloid bait and switch.

Here's how he goes on to report the Eastern European figures:

The number of Poles registering to work fell by a quarter at the end of last year, but arrivals from Latvia and Lithuania more than doubled.
Heh. When the numbers drop they're just people registering to work, but when they rise they're arrivals. See what he did there?

At least he gives us the overall drop from Eastern Europe since 2007 rather than last year. Pat on the back for that one.

After this, we get:

In the year to June 2009, 146,000 British nationals emigrated and 87,000 came back to the UK.
Remember that coverage when those numbers were different? They hit the front page then. They're bunged at the bottom of an article about something else now.
Next, it's this:

This meant that net emigration was 59,000, down from 89,000 in the year to June 2008 – and a peak of well over 100,000 in 2004. In the same time period, net immigration by non-British nationals was 206,000, down from 257,000 in the year to June 2008.
We've been given a figure for the total number of British citizens immgrating back to the UK, but only a net figure for non-British. Do you think that might be because it would contradict the main thrust of the article and ruin the bait and switch, even buried this low?

Right, now we're really, really close to the bottom,it must be time to get on with the asylum seeker figures, so here we go:

Actually, no we don't. There's no mention of that either, aside from the page title, which is probably only still on the website by mistake.

There. Lesson over. When figures aren't all what you want them to be, cherrypick the ones that are, exaggerate them, use them in a bait and switch that covers over some of the figures you don't like that you actually bother mentioning.