Friday, 25 August 2017

Groundhog Day forever with the problem gambling statistics

The Guardian, 2012:

There are an estimated 450,000 "problem gamblers" in the UK, according to the most recent British Gambling Prevalence Survey. And the numbers are rising – up from 0.6% of the population in 2007 to 0.9% in 2010, according to one measure. A further 3.5 million people were categorised as "at-risk" gamblers.

The Guardian, 2015: 

The British Gambling Prevalence Survey 2010 (the last one before its funding was cut) said 900,000 adults were at risk of becoming problem gamblers and 450,000 people admitted they already had a problem.

The Guardian, 2016:

The British gambling prevalence survey indicates that there are around 450,000 pathological gamblers in the country – about 0.9% of the population.

Got that? Is that figure of 450,000 clear enough? That's the official mid-point estimate from the official UK-wide survey of problem gambling in 2010, as repeatedly cited by The Guardian (and other newspapers) for years. (If you are also interested in the more questionable number of 'at-risk' gamblers - see below - that figure is either 900,000 or 3.5 million depending on which hack you ask. Quite a spread, there.)

Now, I'm no Carol Vorderman but if the number of problem gamblers was to rise, it would have to be a bigger number than 450,000, right?

So how do you explain this in today's Guardian?

Number of problem gamblers in the UK rises to more than 400,000

More than 2 million people in the UK are either problem gamblers or at risk of addiction, according to the industry regulator, which warned that the government and industry were not doing enough to tackle the problem.

The report by the Gambling Commission estimated that the number of British over-16s deemed to be problem gamblers had grown by a third in three years, suggesting that about 430,000 people suffer from a serious habit.

A few days ago I issued a plea to people to stop lying about problem gambling being on the rise. I don't know why I bother. See this article from last year if you want to know why this keeps happening.

Instead of checking the facts, the Guardian handed the mic to some anonymous clown from the Campaign for Casino Fairer Gambling - an organisation created and funded by multi-millionaire casino tycoon Derek Webb - to tell its readers what they want to hear:

“The bookies have claimed that because the overall population rate of problem gambling is static, FOBTs are not harmful. The data published today, which shows a rate increase, has totally undermined the bookies’ argument.”

In fact, the Gambling Commission's press release explicitly says that 'overall problem gambling rates in Britain have remained statistically stable' and that the latest estimate of Britain's problem gambling rate is 'similar to the rate published in the 2012 report'.

The Campaign for Fairer Gambling has got nothing to do with making gambling fairer and everything to do with Derek Webb's grudge against the bookies after they put the game he invented onto fixed odds terminal (FOBTs) without giving him a share of the profits ('rather than sue I backed a campaign to make my point'). He will be giggling with glee to see journalists at the Guardian acting as his useful idiots.

But FOBTs are uniquely addictive, right? Not really, not according to today's report from the Gambling Commission:

The lowest rates of problem gambling were found among those who gambled on the National Lottery (1.3%) and the highest were among those who spread bet (20.1%), bet with a betting exchange (16.2%), played poker in pubs or clubs (15.9%), bet on other events with bookmaker (not online) (15.5%) and played machines in bookmakers (11.5%).

The Guardian report didn't mention any of this. Weird, huh?


The Guardian story underwent a a bit of a rewrite before it appeared in print on today's front page. The lie about problem gambling rates rising is still there but it is less prominent and the headline focuses on 'at-risk' gamblers instead.

'At-risk gambler' is one of those terms like 'hazardous drinker', 'overweight' or 'pre-diabetic' which sounds scary but has no clinical significance. Its main purpose is to help campaigners come up with a LARGE NUMBER. (The Sun deserves a dishonourable mention here for interpreting this as 'More than two million people may be problem gamblers'. They're not, otherwise the test would have identified them as such.)

Problem gambling surveys give you a bunch of questions and if you say yes to a certain number of them you are defined as a problem gambler (3 out of 10 for the DSM-IV test and a score of 8 out 27 for the PGSI test - click on the links if you want to do them yourself).

If you say yes to just one of these questions you are defined as a 'low risk' gambler, and being a 'low risk gambler' makes you an 'at-risk' gambler. Say yes to a couple more and you are a 'moderate risk' gambler. Moderate risk gamblers are also at-risk gamblers.

The Guardian tells us that there are two million 'at risk' gamblers in the UK. The implication is that this is a lot (why else would it be on the front page?). In fact, it is a significantly smaller number than has been reported in previous surveys.

In the last UK-wide survey of 2010, 7.3 per cent of respondents were classified as at-risk gamblers. That's about 3.5 million people.

Overall, 5.5% of adults were low risk gamblers (a PGSI score of 1-2) and a further 1.8% were moderate risk gamblers (a PGSI score of 3-7), meaning that overall 7.3% of adults had a PGSI score which categorised them as an ‘at-risk’ gambler. These estimates were similar to those observed in 2007.  

If the report released yesterday is to believed, the percentage has since fallen to 3.9 per cent. That's about two million people.

Overall 2.8% of adults were classed as low risk gamblers (PGSI score of 1 or 2) and a further 1.1% were classed as moderate risk gamblers (PGSI score of 3 to 7). In total, 3.9% of adults had a PGSI score that categorised them as being at-risk gamblers (PGS I score of 1 to 7).

So, in conclusion, we have no change in the number of problem gamblers and a near-halving in the number of at-risk gamblers. Can someone please explain to me why this is front page news?

(Oh yeah, that.)


The Times has managed to outdo the Guardian with its (fairy) story:

Problem gambling grows by 50% in three years

The scale of the gambling epidemic sweeping Britain was laid bare yesterday by official figures showing that 430,000 people in the UK have a serious gambling problem, up from 280,000 in 2012.

A 50 per cent increase in three years has raised pressure on the government to place curbs on the betting industry.

 Times readers with good memories will be confused by this news. In 2011, the paper told them:

It is estimated that there are now between 360,000 and 450,000 problem gamblers in the UK.

 And in February last year they were told:

The last gambling prevalence survey in 2010 found there were 450,000 problem gamblers in Britain but experts at GamCare say the number of addicts is likely to grow in proportion to the size of the industry, which suggests there are now 562,000.

But by October of last year, the 450,000 had doubled to, er, 336,000.

The proportion of people with a severe gambling problem has almost doubled in three years from 0.4 per cent of the population to 0.7 per cent, the equivalent of 336,000 people, according to the Gambling Commission. 

And now it's doubled [sic] again to 430,000. How many times can a number be doubled without getting any bigger?

No comments: