New Analysis Questions Accuracy of UKGC Gambling Survey

A recent study has raised fresh concerns about the reliability of the UK’s primary gambling survey, suggesting that participation numbers may be significantly overstated across a number of activities. The analysis, presented by Regulus Partners researcher Dan Waugh, compares official survey results with industry data and points to some interesting discrepancies. 

New Analysis Suggests GSGB May Overstate Casino and Betting Activity

The Gambling Survey for Great Britain (GSGB) was launched in 2023 as a large-scale online survey of approximately 20,000 participants per year. It has been identified by the UK Gambling Commission as a key benchmark in understanding gambling behavior and the associated risks. However, the latest findings suggest the survey’s estimates may not be indicative of real-world activity in the regulated market.

Waugh’s research involved comparing survey responses with data given by operators on casino visits, soccer pools, and betting exchanges. In each case, the survey found participation figures far higher than those operators had recorded.

Regarding the casino activity, the GSGB indicated that millions of visits were made specifically for table games during a defined period. However, official industry figures showed total visits to casinos were well below that, even before factoring in the number of visitors who do not play table games. This implies that the survey’s estimates might be a few times higher than the real level of participation. 

Study Highlights Survey Bias as Key Factor Behind Gambling Data Gaps

In the case of soccer pools, the divergence was even more marked. The survey data revealed a player base many times larger than what the only operator on the market reported. Betting exchange usage showed a similar tendency with survey estimates, again statistically significantly higher than those based on market data. 

The study accounts for these differences in terms of well-known problems in survey design. One possibility is that people who volunteer to answer gambling-related questionnaires are more likely to gamble than the average person. Other issues are low response rates and sampling imbalances that can bias results. 

Waugh also noted that if the survey numbers were accurate, they would imply that there are large unrecorded gambling markets, which he considers unlikely. Instead, he says, structural biases in the survey are more likely to have caused the discrepancy. 

Findings are relevant to policymakers as GSGB data has been used within discussions on gambling regulation and harm reduction. Regulators should pay more attention to operator data than relying solely on survey-based evidence, Waugh urges. Despite ongoing criticism, the UK Gambling Commission still considers the GSGB to be one of the most comprehensive studies of its kind. However, as it comes under further scrutiny, it is likely to fuel the debate about its accuracy.