Friday, 7 October 2016

The Times' gambling problem

It seems to be that gambling is in the cross-hairs of the wowsers this season. The Times has a bee in its bonnet about it (along with sugar) and so runs front page stories like this...

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.

As I have tried to explain many times, almost everything people think they know about gambling in Britain is wrong. That is hardly surprising when anti-gambling folk spend all their time trying to mislead them, but there really is no excuse for this.

The Gambling Commission does not say what The Times claims. Not even close. Surveys of problem gambling were carried out in the UK in 1999, 2000 and 2010, with a further survey of England in 2012 and Scotland in 2013. There has been no survey since. The Gambling Commission published its latest report on the matter in July 2014. It was comprehensive and concluded the following:

Overall, the rates of past year gambling reported in the combined health survey series [of 2012-13 - CJS] are typically lower than those reported in the BGPS series [of 1999-2010 - CJS]. Results from this present health surveys report showed that 65% of adults had gambled in the past year, whereas estimates from the BPGS series ranged from 72% in 1999, to 68% in 2007 to 73% in 2010.

According to the combined health survey data, the problem gambling rate as measured by the DSM - IV was 0.5%. This was similar to problem gambling rates observed in the BGPS series which for England and Scotland were 0.6% in both 2007 and 1999 and 0.9% in 2010. The differences between survey years were not significant.

Problem gambling rates according to the PGSI were also similar between the surveys, being 0.4% for the combined health survey and 0.6% in BGPS 2007, and 0.7% in BGPS 2010.

Rates of problem gambling according to either the DSM-IV or PGSI did vary by survey year. Estimates were highest in 2010 (1.2%) and were lower in both the BGPS 2007 (0.8%) and the combined health survey data (0.6%).

Overall, problem gambling rates in Britain appear to be relatively stable, though we caution readers against viewing the combined health survey results as a continuation of the BGPS time series. This is because of the change of survey vehicle which could affect our ability to make direct comparisons.

So there are fewer people gambling and no change in the rate of problem gambling. The rise of online gambling, the introduction of fixed-odds betting terminals and a dramatic increase in gambling advertising has had no effect on rates of problem gambling.

The rates recorded in each survey for England and Scotland (combined) (with two different methodologies used in each instance) are as follows:

You would have to be on drugs to interpret this as problem gambling 'almost doubling'. Indeed, a literal interpretation of the figures from 2010 and 2012 would suggest that it has almost halved.

That would be wrong, however, because these are only estimates - albeit from a large sample - and the differences between survey years are not statistically significant.

But The Times is wrong on so many levels (as is The Sun). There are no data for the last three years. Neither methodology is designed to identify a 'severe gambling problem'. Such data as there is suggests no change in problem gambling prevalence and the Gambling Commission has said so.

The next survey must be imminent but even if its shows a rate of 0.7 per cent, it wouldn't indicate that problem gambling has 'almost doubled'. It would be, at best, regression to the mean.

This isn't the first time Britain's newspaper of record has printed complete bollocks about gambling on its front page. In today's edition there is even an op-ed from casino tycoon Derek Webb of Prime Table Games (trading as the Campaign for Fairer Gambling) demanding a clampdown on gambling machines in bookies.

No comments: