Gambling on Google and Twitter: Harnessing Big Data and Machine Learning to Understand the Impact of COVID-19 on Gambling Harm

Session Title

Session 3-4-E: The Effects of COVID-19 on Gambling

Presentation Type

Paper Presentation

Location

Park MGM, Las Vegas, NV

Start Date

25-5-2023 3:30 PM

End Date

25-5-2023 5:00 PM

Disciplines

Marketing | Psychology | Public Health

Abstract

Abstract: Gambling behaviour has certainly been impacted by the COVID-19 pandemic, but little is known about how the gambling industry responded or how online search-related behaviours may have changed across years. Using contemporary ‘big data’ approaches across two studies, we tracked gambling-related Google searches and social media (Twitter) use by gambling industry operators before, during and since COVID-19 restrictions in the UK. We examined how people searched for gambling opportunities and compared this to publicly available operator data. We also examined tweets from gambling operators and affiliates to identify trends in the frequency of tweets and the emotionality of language used. Findings demonstrate that operators have proven remarkably agile in their social media use during and since the pandemic, and that trends in Google searches for gambling were associated with both the non-availability of traditional forms of sports betting and industry participation and advertising data. Overall, for the first time, the interconnectedness of our big data approach has identified meaningful trends on the impact of COVID-19 upon both individual data and industry marketing strategies.

Implication statement: The findings of this project have highlighted the potential utility of using online search behaviour as a proxy measure of gambling-related harm. They also provide a comprehensive understanding of industry social media marketing strategies throughout the pandemic and therefore will be of interest to both academics and policymakers.

Keywords

Gambling marketing, COVID-19, Machine learning, Sports Betting, Sentiment Analysis, Time-series data

Author Bios

Dr Scott Houghton is a lecturer at Northumbria University and worked as a research assistant at Swansea University on the Gambling on Google and Twitter project. His PhD research explored how gambling was marketed on social media and how this impacted upon gambling behaviour. His PhD research was funded by GambleAware.

Dr Frederic Boy (PhD) is an assistant professor at Swansea University. He is a multidisciplinary scientist (Data Science, Psychology, Neuroscience) educated in Grenoble University (BSc, MSc), University of Edinburgh (MSc) and the CNRS (PhD). During his postdoctoral training at the world-class Cardiff University Brain Research Centre (CUBRIC), he was funded by the Wellcome Trust and part of the team that developed one of the first technique to image functional chemistry in the living human brain.

Dr Alexander Bradley is a senior lecturer at the University of Portsmouth who completed his PhD at the University of Nottingham. During his PhD Alex completed an internship in the Parliamentary Office of Science and Technology researching the effects of behaviour change policies. Alex has carried out a range of projects exploring the strategies gambling companies use on social media to advertise their products and utilising big data methods to collect novel interesting data.

Dr Richard James is an assistant Professor in the School of Psychology at the University of Nottingham. His research has primarily looked at the behavioral mechanisms involved in mobile gambling and behavioural addiction, and the latent structure of disordered gambling and other disorders.

Dr Heather Wardle is a social scientist with nearly twenty year's experience designing, implementing and analysing some of Britain's largest studies of health and wellbeing, including the Health Survey for England and the British Gambling Prevalence Survey. She is a specialist in gambling research, policy and practice and is currently leading the Lancet Public Health Commission on Gambling. Her research focuses on understanding the impact of gambling on people's lives and devising solutions.

Simon Dymond is Professor of Psychology and Behaviour Analysis at Swansea University, where he is Director of the Gambling Research, Education and Treatment (GREAT) Network Wales, Co-Director of the Experimental Psychopathology Lab, and Visiting Professor at Reykjavík University, Iceland. He has received funding from a diverse range of international funders, such as Health and Care Research Wales, GambleAware, and the International Center for Responsible Gaming.

Funding Sources

The work described here was supported by an award from the British Academy/Leverhulme Small Research Grants Scheme (SG2122\211340).

Competing Interests

SH previously received funding from GambleAware, who receive voluntary donations from the gambling industry, to carry out my PhD studies. RJEJ is currently supported by research grants from the Academic Forum for the Study of Gambling. The funds for this research were received from regulatory settlements made by gambling companies. RJEJ was previously co-investigator on a project funded by the International Centre for Responsible Gaming, which is a charitable organisation funded by donations from the gambling industry. In the last three years, HW has worked on one project funded by GambleAware, looking at gambling and suicidality. Between 2015 and 2020 she was Deputy Chair of the Advisory Board for Safer Gambling, providing independent advice to government on gambling policy. She was remunerated by the Gambling Commission (the regulator). The other authors declare no competing interests.

Share

COinS
 
May 25th, 3:30 PM May 25th, 5:00 PM

Gambling on Google and Twitter: Harnessing Big Data and Machine Learning to Understand the Impact of COVID-19 on Gambling Harm

Park MGM, Las Vegas, NV

Abstract: Gambling behaviour has certainly been impacted by the COVID-19 pandemic, but little is known about how the gambling industry responded or how online search-related behaviours may have changed across years. Using contemporary ‘big data’ approaches across two studies, we tracked gambling-related Google searches and social media (Twitter) use by gambling industry operators before, during and since COVID-19 restrictions in the UK. We examined how people searched for gambling opportunities and compared this to publicly available operator data. We also examined tweets from gambling operators and affiliates to identify trends in the frequency of tweets and the emotionality of language used. Findings demonstrate that operators have proven remarkably agile in their social media use during and since the pandemic, and that trends in Google searches for gambling were associated with both the non-availability of traditional forms of sports betting and industry participation and advertising data. Overall, for the first time, the interconnectedness of our big data approach has identified meaningful trends on the impact of COVID-19 upon both individual data and industry marketing strategies.

Implication statement: The findings of this project have highlighted the potential utility of using online search behaviour as a proxy measure of gambling-related harm. They also provide a comprehensive understanding of industry social media marketing strategies throughout the pandemic and therefore will be of interest to both academics and policymakers.