Abstract:
The Good Behaviour Game (GBG) (Barrish, Saunders & Wolf, 1969) is an empiricallytested, effective procedure to reduce disruptive and challenging classroom behaviour. It is internationally acclaimed (although little-known in New Zealand) for its preventive properties for reduced risk of early-adulthood antisocial disorders, criminal convictions and drug and substance abuse. Whilst it has been shown to be effective in a wide range of applications and settings, and with various populations, the original GBG utilises negatively-stated behavioural rules and a mildly aversive contingency, both of which are in conflict with positive behavioural psychology practices. The present study compares a novel, previously untested variant (GBG2) with the original GBG (GBG1), that retains the negatively-stated rules but adopts a positive contingency, to evaluate relative effectiveness in the reduction of disruptive classroom behaviour. An aim of the study is to introduce the GBG2 into the NZ education setting; it is hoped that GBG2 may eventually be considered for inclusion in the Ministry of Education’s school-wide Positive Behaviour for Learning (SWPB4L) initiative. In the present study, both GBG variants were implemented in two general education primary school classrooms in an alternating treatments design with changing criteria. GBG2 was found to be more effective in reducing disruptive behaviour than GBG1 overall, and demonstrated greater reductive effects in the younger (six-year-old) class than with the older children (eight-yearolds). Children of both age-groups preferred GBG2 over GBG1 as it was easier to play due to the positive focus, and both teachers cited the positive ideology of GBG2 as being the most important consideration when deciding preference between the two variants. Both teachers and children recommended its use in the wider school. Future research recommendations include a direct comparison study of GBG2 with another variant with explicit, positivelystated rules and positive contingencies, to isolate and evaluate the effects of the rules component.