Consultancy Regulus Partners says the Commission weakened key guidance on its Gambling Survey for Great Britain (GSGB) at a politically sensitive time. The impact of the relaxed guidance was that decision makers, and the public as well, found it easier to consider the survey a valid measure of gambling-related harms within the nation.
Why the Guidance Shift Matters
From the very outset, the GSGB has been controversial due to the high estimates of gambling prevalence and problem gambling among the population compared to previous British surveys. The figures associated with the study suggested that 48% of the adult population of Great Britain had gambled within the last month, while 2.7% had a PGSI score of 8 or more.
Earlier guidance had urged caution around using PGSI-based estimates at the population level. However, the Commission later updated its guidance in October 2025 and explicitly allowed population expressions for GSGB estimates.
As Regulus observes, the warning was subsequently undermined ahead of the Autumn 2025 Budget, which later introduced major duty changes for parts of the gambling sector. The consultancy claims that the reinterpretation of the policy added political weight due to the timing of events, either intentionally or not.
What the Study and FOI Material Suggest
One of the key studies published in August 2025 by researchers from the London School of Economics (LSE) and NatCen aimed at explaining the reason why the GSGB had been generating higher estimates compared to other surveys, such as the Health Survey for England. According to Regulus, however, the results were mixed.
Its interpretation of the results from the three tests is rather straightforward. The first seemed to favor the claim that there is a higher chance of responding to the survey if its topic relates to gambling, thus distorting the statistics on survey participation rates. Another test revealed certain discrepancies between responses made via phone and online. Regulus, however, argues that these findings have little relevance when compared to the GSGB and NHS surveys that use self-completion questionnaires for gambling questions. A third test failed to support the notion that a longer list of activities was responsible for the discrepancy.
The most delicate aspect of the report is the material obtained through freedom of information. Regulus cites an August 2025 email in which an unnamed LSE researcher reportedly said the experimental results did not justify changing the guidance on ‘grossing up’. The report also refers to internal review comments asking if there was a sufficient basis for lifting the restriction.
A Wider Test of Statistical Trust
The Gambling Commission rejects this criticism and maintains its support for the GSGB and the associated guidance. The Commission also highlights that the survey has received considerable external review.
But the controversy is no longer limited to one dataset. Regulus has called for an independent review and asked the Office for Statistics Regulation to consider its findings as part of its examination of the GSGB.
The episode illustrates how disputed survey evidence can become more consequential when it sits close to tax policy. In such a case, even a slight wording change can have far-reaching implications.
