Report
Illegal online gambling - Phase 3: Understanding push factors for consumer engagement with illegal gambling websites
The Gambling Commission's research into consumer engagement with the illegal online gambling market.
Limitations of the approach
Despite the strengths of the work, these research methodologies also each have their own limitations. Findings should therefore be read as a triangulated account of the overall story rather than as a single definitive measure.
Digital intelligence
Digital intelligence provides insight into public social media content and conversation and can help identify behavioural trends and patterns. However, it reflects expressed rather than confirmed behaviour and cannot be treated as representative of the wider population. Digital intelligence naturally tends to capture those who are more vocal or hold stronger opinions on subjects. For example, individuals who feel strongly about certain regulatory measures, are often more vocal in digital spaces, whereas those with more neutral feelings may be less likely to post. Furthermore, mentions of intent can’t reliably verify actual migration to illegal online gambling, or provide a robust quantitative sizing of the prevalence of the activity.
The analysis is also limited to publicly available content. Because of the sensitive nature of the topic, the anonymity provided by platforms like Reddit allows space for those who illegally gamble to share their experiences more honestly. On social media platforms like X and Facebook, discussion was more limited, and while YouTube contained a high volume of promotional videos, it lacked genuine conversations between those who gamble. This method therefore cannot capture the full extent of activity taking place, and instead sets up measurement for the quantitative phase.
Quantitative research
Quantitative research provides a broad view of attitudes and behaviour across a defined sample. However, it is less adept on its own explain why participants hold particular views or what factors shape their behaviour.
As outlined in the methodology section, this research focuses on those who have gambled online. The findings should therefore not be treated as representative of all those who gamble. Caution should also be taken when interpreting subgroup differences, such as those relating to gambling frequency or Problem Gambling Severity Index (PGSI) score, as these groups may not themselves be representative, given the research was conducted with an online panel, which naturally over-index for online gambling frequency and PGSI compared to the full population, where online engagement is more varied.
Qualitative Research
By their nature, qualitative sample sizes are small and exploratory. Therefore, qualitative findings should be treated as illustrative and directional unless explicitly contextualised by the quantitative data. Furthermore, qualitative research relies on interpretation, introducing the possibility of researcher bias. This was mitigated through a grounded theory approach and the systematic analysis of all focus group and pre-task data and notes.
Qualitative participants were recruited directly from the quantitative survey. Due to natural drop-off, including participant availability and willingness to engage, the qualitative focus groups consisted of 15 participants in total: 7 Skilled Advocates, 4 Social Explorers, and 4 Accidental Tourists. Because of the limited sample size for the latter 2 groups, Social Explorers and Accidental Tourists, this limited our ability to fully unpick qualitative differences in attitude between the 2 cohorts. Readers should keep this in mind, as the quantitative data analyses these segments separately. To compensate, quantitative findings have been woven throughout the qualitative reporting to ensure distinct insights for all audiences are accurately reflected.
It should also be noted that while the quantitative phase provided insight into Self-Excluders as a distinct audience, this group was deliberately excluded from the qualitative phase of the research. This was a necessary safeguarding measure to avoid any potential distress or discomfort associated with discussing gambling-related harm in a live group setting. It was also due to previous phases of research providing an existing understanding of the Self-Excluder group’s motivations and pathways into illegal online gambling.
Qualitative insights gathered in this study are based on self-reported, rather than observed, behaviour. Consequently, findings are limited to what participants felt comfortable disclosing. Because group dynamics can sometimes inhibit a participant's willingness to share sensitive behaviours or perceptions, independent pre-tasks were deliberately included in the qualitative design. This enabled the qualitative phase to capture private recollections without the influence of the group dynamic. However, reliance on self-reporting remains an important caveat to keep in mind.
Previous sectionStrengths of the approach Next section
Ethics and governance
Last updated: 20 October 2026
Show updates to this content
No changes to show.