International Scientific Evidence

Key studies by Dr. Sheppard Kellam and colleagues were part of a long-term research partnership between the Baltimore City public school system and Johns Hopkins University, (see Kellam, Reid & Balster, 2008 for an overview). From three generations of randomised controlled field studies in the US, it was found that the GBG had a significant and meaningful
impact on young people, especially males, from childhood through to young adulthood (aged 19-21).

Educational progress and attainment

Research from the Baltimore studies, and also from Belgium and the Netherlands, has shown that the GBG can reduce disruptive and aggressive behaviour in classrooms and improve children’s ability to focus and produce more work. Improvements in team work and social skills were also observed. A range of studies, of varying scientific quality, have specifically reported GBG impacts on educational progress and attainment:

US studies, range of scientific quality:

  • Decreased disruptive behaviour (Barrish, Saunders, & Wolf, 1969;  Lannie & McCurdy, 2007)
  • Decreased off-task behavior of disruptive adolescents (12-13 yrs)  (Phillips & Christie, 1986)
  • Increased rates of completed math assignments (Darveaux, 1984)
  • Increased math skills (Harris & Sherman, 1973)
  • Improved creative writing skills (Maloney & Hopkins, 1973)
  • Increased student productivity (Lutzer & White-Blackburn, 1979; Robertshaw & Hiebert, 1973)

US studies, high quality randomised trials:

  • GBG children showed improvements in educational attainment. For example, compared with a control group, GBG participants scored higher in reading and in maths with gains approximately equivalent to one additional year of academic progress. They were 36% less likely to have received special education services in primary and secondary school, 21% more likely to have completed their secondary education and 62% more likely to be attending university (Kellam et al., 2008; Kellam et al., 2011)

Canadian study, high quality randomised trial:

  • Students in Good Behaviour Game classrooms showed greater levels of attentiveness and literacy skills compared with controls (Dion et al., 2011)

Dutch & Belgium studies, high quality randomised trials:

  • Increase in on-task behaviour and decrease in talking-out behaviour (Leflot et al., 2010)
  • Improvement in spelling scores among GBG children, controlling for initial level of spelling in second grade.  Results are similar for boys and girls (Craeyevelt et al., 2012)

Longer-term outcomes

Over the longer-term and into young adulthood, participation in GBG classrooms led to a reduction in bullying behaviour, antisocial and violent behaviour, criminal activity, drug and alcohol problems and mental health problems, particularly with those children who are most at risk. The Baltimore studies have shown significant and important reductions in a broad set of problem outcomes into early adulthood (i.e. ages 19-21), for example:

  • 50% lower rates of lifetime illicit drug abuse/dependence
  • 59% less likely to smoke 10 or more cigarettes per day
  • 35% lower rate of lifetime alcohol abuse/dependence.

These findings were particularly strong for males who, at an early age, were identified as more aggressive or disruptive: amongst the boys identified as most at risk, 29% who played the GBG reported drug use disorders by early adulthood, compared with 83% of controls. (Kellam et al., 2008; Kellam et al., 2011)

Cost Effectiveness

Aos et al. (2004) investigated the medium to longer-term potential benefits of the GBG for individuals and broader society in the US. The GBG returned $25.92 per $1.00 spent on the programme (for just some outcomes). The average saving per pupil was $196, and the GBG was the least costly to implement of six evidence-based child development interventions, had the second best cost effectiveness ratio, and had the strongest evidence base over the longer-term.

In a recent update to the 2004 cost-benefit analysis, Aos et al (2011) report that, taking account of the most recent scientific evidence, the GBG now shows a return of $96 for every $1 spent on the programme.