GambleAware research: Online surveys “over-estimate” UK gambling harm
Charity-commissioned study highlights ‘selection bias’ in targeting online and frequent gamblers over paper questionnaires and in-person interviews
New research commissioned by GambleAware has suggested conducting online-only surveys of UK gamblers could lead to skewed and “over-estimated” data on problem gambling.
The study, carried out by London School of Economics professors Patrick Sturgis and Jouni Kuha, looked at how methodological differences between surveys affect the accuracy of estimates of gambling harm.
The project was launched after a 2019 YouGov survey reported substantially higher rates of gambling harm than had previously been estimated by the 2018 Health Survey for England, which used the Problem Gambling Severity Index (PGSI).
The discrepancy led to concerns about the true levels of gambling harm within the adult population of the UK and GambleAware saw fit to commission this latest project.
Researchers found that predominantly or entirely online self-completion surveys produced consistently higher estimates of gambling harm than paper questionnaires and face-to-face interviews.
Other potential causes of this disparity, such as true change over time, differences in population coverage and differential measurement errors were ruled out by researchers.
“These discrepancies in sample composition seem likely to be driving the differences we observe in rates of gambling harm,” the report stated.
“Through processes of self-selection, the online surveys skew toward people who are more online and ‘tech savvy’ and, perhaps unsurprisingly, these sorts of people are more likely to be online gamblers compared to people who gamble in person.
“If samples contain disproportionate numbers of online and frequent gamblers (compared to the general population), surveys will tend to over-estimate gambling harm because, as we were able to demonstrate, online and frequent gambling are independently associated with higher probabilities of gambling harm,” the report added.
The professors have suggested online surveys should still continue due to their lower cost and that measurement of gambling-related harm should move entirely online in the future.
“The move to online interviewing should be combined with a programme of methodological testing and development focusing in particular on strategies for mitigating selection bias and on improving the accuracy of measurement of gambling harm,” the report added.
However, researchers recommended that in-person surveillance of players should continue on a regular but infrequent basis in order to confirm estimates from online surveys and to provide an ongoing benchmark for evaluating the effectiveness of that research.
Addressing the report’s findings, GambleAware research director Alison Claire said: “We want our prevention, treatment and support commissioning to be informed by the best available evidence, and having survey data we can be confident in, within the constraints of data collection in an increasingly online world, is key.
“GambleAware’s annual survey is an important tool in building a picture of the stated demand for gambling harms support and treatment, and of the services, capacity and capability needed across Great Britain to meet that demand,” Claire added.