One such set of figures appeared in Sunday’s News of the World, which ran a piece on the ‘jobless ghettoes’ of the UK, where the amount of unemployed people runs at far higher levels than the national average.

But just how have these figures been put together?

The Claim

According to the red top, uppermost on this list of jobless blackspots was the Welsh town of Merthyr Tydfil.

The report, written by Chief Political Correspondent David Wooding, claimed: “The league table of despair is topped by the former steel and coal town of Merthyr Tydfil in South Wales, where 30.1 per cent are unemployed.”

So what do the stats say?

Analysis

When Full Fact contacted Mr Wooding, he confirmed that his figures were those published last week by the Office for National Statistics (ONS).
 

But using these figures it is hard to find a number to match to that used by the News of the World.

The figures for Merthyr Tydfil show the unemployment rate for the town as 12.6 per cent – far lower than the figure quoted in the paper, but this itself is no reason to question the stats in the report due to the different ways in which one could count people who are unemployed.

For instance, the rate for people who were recorded as economically inactive is 27.8 per cent. Still lower than the News of the World figure.

Likewise, if the figure for unemployment is the measure of those claiming ‘key out of work benefits’, namely job seekers, incapacity benefits, lone parents and others on income related benefits, the rate is 25.2 per cent.

The total proportion of benefit claimants is 28.8 per cent – again still short of the elusive 30.1 per cent.

Struggling to find the figure, we contacted the ONS, who after some number crunching, informed us that the only way they could get the numbers to add up was by using slightly out of date calculations.

It appears that the NotW figures take the total number of people claiming benefits in Merthyr, and divide it by the working age population.

The problem is they seem to be using a different number for the town’s working age residents to that which is currently used. As the statisticians explained, working age population of an area used to be men aged 16-64 and women aged 16-59.

The ONS now use a population figure including both men and women between 16 and 64, and though last week’s stats were the first to use this measure, warning of the change was sent out months in advance.

Because a lower figure is used for the population, this means the proportion of people on benefits is actually higher than official figures suggest.

Though we have not been able to confirm this was the way the numbers used in the report were produced, such calculations, when used for the other towns mentioned in the story, produced percentages only slightly different those the paper published.

If this was indeed how the NotW figures were calculated, this also raises a further question about the stats. As well as total benefit claimants as a proportion of the population, the ONS data includes a further breakdown of the amount claiming key out of work benefits.

As an ONS statistician explained, some of the benefits considered in the total claimants figure could still be received by people who were actually working, so are not counted as key out of work benefits.

The table below shows the extent to which both ignoring changes to the statistical methods, and including all benefit claimants serves to inflate the figure used by the News of the World for a number of areas.
 

However the statistics seen by Full Fact suggest that the report in the News of the World used figures that are higher than the published official data.

Although there could be another explanation for the differences, it seems the numbers include people claiming benefits who may be working, and exclude some women who are now classed as working age.

So while the paper may have been justified to warn about the numbers of people out of work, it seems they weren’t completely on the money when it came to their own working out.

Patrick Casey