The UK Gambling Commission has released a new study aimed at explaining why different surveys show significant differences in gambling participation rates and the Problem Gambling Severity Index (PGSI). The study, led by Professor Patrick Sturgis, focuses on how survey design influences data outcomes.
The study found that self-administered surveys like the Great Britain Gambling Survey (GSGB) typically report higher rates of gambling participation, whereas face-to-face surveys like the Health Survey for England (HSE) or the Adult Psychiatric Morbidity Survey (APMS) report lower figures. The study explored three factors: mentioning gambling in the survey invitation, whether respondents were interviewed by an interviewer, and the impact of an updated, longer list of gambling activities on the results.
The results show that explicitly mentioning gambling in the survey invitation does not affect the overall response rate, but does increase the reported gambling participation rate by about 4 percentage points, indicating that those interested in gambling are more likely to participate in the survey. The proportion of people with a PGSI score of 1 or above was slightly higher by 1.8 percentage points in this group, but the difference was not statistically significant.
Furthermore, telephone interviews significantly suppressed respondents' reports of adverse behaviors. Under online self-administration, the proportion of respondents with a PGSI score of 1 or above was 4.4 percentage points higher than in telephone interviews, equivalent to nearly a 50% increase in the number of respondents reporting problem gambling behaviors. Updating the list of gambling activities had little impact on reported participation rates or PGSI scores.
The study suggests that regulatory bodies update the interpretation guidelines for GSGB data to more accurately reflect the differences between survey methods. Ben Harden, Director of Research and Policy at the Gambling Commission, stated that the study provides important insights into the impact of different survey methods and strengthens confidence in the GSGB results.
Professor Sturgis noted that while it is impossible to determine the "true" rate of gambling participation, the experimental results help understand how different survey designs affect outcomes, providing a basis for future improvements in data collection and interpretation.
