Three experiments on the causes of differences in estimates of gambling and gambling impacts in general population surveys

This research report examines the causes of differences in estimates of gambling and its impacts in survey data. The study looked into the effects of survey wording that explicitly mentions gambling in the invitation; the presence of an interviewer; and the length and specificity of the list of gambling activities in the survey. These aspects were considered in relation to whether respondents reported gambling in the past year and a score 1 or above on the Problem Gambling Severity Index (PGSI). The experimental design of the study provides stronger evidence than previous observational studies that how the topic is framed and the way the survey is given influence estimates of gambling and the impacts of gambling.

Findings include:

• Naming gambling explicitly in the survey invitation did not affect the overall response rate, but led to a slight increase in survey participation from people who had an existing interest in gambling. 
• The presence of an interviewer asking the survey questions had a substantial effect on self-reported gambling impacts, which suggests that people reported less behaviour that was socially undesireable when an interviewer was present. 
• Including newer formats of gambling in the survey led to slightly higher reported participation in gambling, but did not change PGSI estimates, which suggests that this did not not account for the difference between previous and more recent survey estimates.
• Recommendations include adjustment of the Gambling Survey for Great Britain’s estimates of gambling and gambling impacts to take into account potential differences from past estimates.