Relationship of Dosage and Fidelity of the Good Behavior Game with Teacher Efficacy and Burnout

Introduction: Although research has shown positive associations between implementation of school-based interventions, as measured by dosage, adherence, and fidelity, and student outcomes (Durlak & DuPre, 2008), less is known about how level of implementation relates to implementers’ outcomes. Given the competing demands for time and money in schools, as well as concerns regarding staff turnover and job satisfaction (Marvel et al., 2006), it is important to determine whether additional time and effort devoted to implementing evidence-based interventions is positively or negatively related to teachers’ personal resources. This study aimed to examine the relationship between Good Behavior Game (Embry et al., 2003) dosage (i.e., amount of time spent playing the games) and fidelity (i.e., measured by external observers) and teacher-reported burnout and efficacy across a school year.  

Method: Data come from 199 teachers in 18 schools involved in an RCT testing GBG versus an integration of GBG with the PATHS (Greenberg & Kusché, 2011) social emotional curriculum. Teacher-reported efficacy was measured on two scales: behavioral management (Main & Hammond, 2008) and social emotional learning (SEL; Domitrovitch & Poduska, 2008). Burnout was measured using the three scales (emotional exhaustion, depersonalization, personal accomplishment) on the Maslach Burnout Inventory (Maslach et al., 1997).  The targeted predictor variables were observational ratings of implementation fidelity and the number of minutes played per game across the school year. Teacher and school characteristics were included as controls. Hierarchical linear modeling accounted for the repeated nature of the outcome variables (i.e., measured at four time points across the year) and nesting of teachers within schools (Raudenbush & Bryk, 2002).

Results: Preliminary analyses demonstrated that the average implementation score across four observations was significantly related to change in depersonalization over time, such that teachers who demonstrated a high level of fidelity showed increasing levels of burnout across the school year (b = .13, SE = .06, p < .05). More time spent playing GBG was also related to intercept, but not slope, of efficacy for behavioral management (b = .02, SE = .01, p < .05) and SEL (b = .02, SE = .01, p < .05).

Discussion: These results suggest that teachers who spent more time implementing this evidence-based intervention reported feeling more efficacious, but that better fidelity was associated with increased burnout over time. Implications for prevention and implementation science will be discussed.