Close
Close
Browser Compatibility Notification
It appears you are trying to access this site using an outdated browser. As a result, parts of the site may not function properly for you. We recommend updating your browser to its most recent version at your earliest convenience.
Skip to Content
Home
I'd Like To...
Gambling Research Exchange Ontario Logo
Contact Us Main menu icon
  • About Us
    • What We Do
    • Team
    • Board of Directors
    • Join Us
    View our Evidence Centre page
    Search the Evidence Centre
  • Services
    • Applied Research
    • Knowledge Products 
    • Knowledge Management
    • Stakeholder Engagement
    • Impact Evaluation
    • Project Consulting
    View our Evidence Centre search page
    Search the Evidence Centre
  • Resources
    • Conceptual Framework of Harmful Gambling
    • Data Repository
    • Evidence Centre
    • Gambling from a Public Health Perspective
    • Research to Inform Action Evidence Hub
    • Safer Gambling Evaluation Evidence Hub
    • Resources for Safer Gambling During COVID-19
    View our Evidence Centre search page
    Search the Evidence Centre
  • Partners
    • Gambling Commission
    • Funding Opportunities
    View our Evidence Centre page
    Search the Evidence Centre
  • Contact
    • Get in Touch
    • Helplines
    View our Evidence Centre page
    Search the Evidence Centre
  • Search
News:
Pause

Predicting self-exclusion using machine learning

Show or hide navigation More
Decrease text size Default text size Increase text size
Print This Page
Share This Page
  • Open new window to share this page via Facebook Facebook
  • Open new window to share this page via LinkedIn LinkedIn
  • Open new window to share this page via Twitter Twitter
  • Email This page Email
View Abstract View Snapshot Back to Search Results

snapshot summaries


Author(s): Finkenwirth, Stephanie ; MacDonald, Kent I. ; Deng, X. ; Lesch, T. ; Clark, Luke

Journal: International Gambling Studies

Year Published: 2020

Date Added: February 17, 2021

People who gamble online may experience more gambling problems. This could be due to the easy availability of online gambling. However, online gambling allows for people's gambling behaviour to be tracked over time. This is called behavioural tracking. In this study, the researchers used behavioural tracking data to predict voluntary self exclusion (VSE), which might be a sign of problem gambling. The researchers obtained data from a provincially owned online gambling site in Canada. The de-identified data set contained 2,157 players who had a record of VSE and 17,526 who did not.

The researchers selected 20 variables as predictors of VSE. The variables included measures of gambling frequency, intensity, and variability. Overall, the variables predicted VSE well. The most important predictor of whether a player had a record of VSE was variability in the amount of money bet per session. It captured unusual betting patterns that fluctuated a lot in the amount of money bet. Another important predictor was Bets per Day. Variance in Money Bet per Session and Bets per Day together accounted for 58% of the predictive signal. Overall, behavioural tracking data could be used to identify and intervene with people experiencing problem gambling.


Citation: Finkenwirth, S., MacDonald, K., Deng, X., Lesch, T., & Clark, L. (2020). Using machine learning to predict self-exclusion status in online gamblers on the PlayNow.com platform in British Columbia. International Gambling Studies. Advance online publication. https://doi.org/10.1080/14459795.2020.1832132

Article DOI: https://doi.org/10.1080/14459795.2020.1832132

Keywords: chasing losses ; gambling disorder ; machine learning ; online gambling ; slot-machine gambling

Topics: Online Gambling

Conceptual Framework Factors:   Exposure - Gambling Setting ; Types - Structural Characteristics ; Exposure - Accessibility ; Resources - Harm Reduction, Prevention, and Protection ; Psychological - Judgement and Decision Making ; Gambling Exposure ; Gambling Resources ; Resources - Interventions

Study Design: Other Quantitative

Geographic Coverage: Canada, British Columbia

Study Population: N = 19,683 players of the eCasino section of PlayNow.com who made at least 200 bets between October 1, 2014 and September 30, 2015

Sampling Procedure: Data were provided by the British Columbia Lottery Corporation in a de-identified format

Study Funding:

This study was funded by a Research Grant from the Province of British Columbia Ministry of Finance (Gaming Policy & Enforcement Branch).

Login to Edit

Receive Email Updates...
×

© 2021 Gambling Research Exchange

55 Wyndham St. N., Suite 214A
Guelph, ON, Canada, N1H 7T8
Tel: (519) 763-8049

Twitter icon

AccessibilityPrivacySitemapEvidence CentreContact UsBoard Login
Designed by eSolutions
  • About Us
    • What We Do
    • Team
    • Board of Directors
    • Join Us
  • Services
    • Applied Research
    • Knowledge Products 
    • Knowledge Management
    • Stakeholder Engagement
    • Impact Evaluation
    • Project Consulting
  • Resources
    • Conceptual Framework of Harmful Gambling
    • Data Repository
    • Evidence Centre
    • Gambling from a Public Health Perspective
    • Research to Inform Action Evidence Hub
    • Safer Gambling Evaluation Evidence Hub
    • Resources for Safer Gambling During COVID-19
  • Partners
    • Gambling Commission
    • Funding Opportunities
  • Contact
    • Get in Touch
    • Helplines