BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:What can gambling machine data tell us about betting behaviour? - 
 David Excell
DTSTART:20170224T110000Z
DTEND:20170224T120000Z
UID:TALK71315@talks.cam.ac.uk
CONTACT:Anita Faul
DESCRIPTION:Bookmakers’ betting machines continually attract political a
 nd media\nattention over their impact on problem gambling. Featurespace an
 d NatCen\npartnered with the Responsible Gambling Trust to investigate har
 mful\npatterns of play on gaming machines\, and draw implications for\nint
 ervention. The methods and results of this ground-breaking investigation\n
 are presented\, linking industry-held data from the five largest UK\nbookm
 akers with surveys of loyalty card customers (which measured\ncustomers’
  Problem Gambling Severity Index score as a proxy for harmful\nplay)\, int
 egrating research methods for 10 billion individual gaming\nmachine events
 . The data set included 6.7 billion bets and 333\,000\ncustomers.\n\nThis 
 huge data set was harnessed to model actual gaming play\, measure\ntheoret
 ical markers of harm (e.g. faster gaming)\, survey loyalty card\ncustomers
  (matching 4\,001 responses with transactional data) and explore\nconsumer
  interventions.\n\nTwo predictive models were built\, exploring the statis
 tical relationships\nbetween the data and the customer surveys. The result
 s showed it is\npossible to distinguish between problem gamblers and non-p
 roblem gamblers\nin industry data:\n\n Player model: behavioural analys
 es in loyalty card holder data – 66%\nimprovement over the current basel
 ine model.\n\n Session model: proxy measurements for anonymous players 
 rather than\nindividual players –550% improvement in accuracy of detecti
 ng problem\ngamblers over the industry standard.\n\nThe research demonstra
 ted that a combination of variables are needed to\nidentify problem gamble
 rs\, in contrast to proposed policy suggestions of\nregulating individual 
 parameters (e.g. stake size). Approach limitations\nare assessed\, includi
 ng data skewedness\, and explore the challenges of\nincorporating big data
  into social scientific investigations.\n
LOCATION:Rayleigh seminar room\, 2nd floor\, Maxwell Building\, Cavendish 
 laboratory
END:VEVENT
END:VCALENDAR