![[ Visit Client Website ]](images/banner.gif)
The Good Behavior Game (Barrish et al., 1969) and Promoting Alternative Thinking Strategies (PATHS; Greenberg et al., 1995) are two universal school-based preventive interventions which have been shown in randomized controlled trials (RCTs) to have an immediate and beneficial impact on aggressive/disruptive and off-task behavior. The current study aimed to test the effects of the PAX version of GBG when combined with the PATHS program, relative to PAX GBG alone and a control condition. We expected additive, if not synergistic, effects as a result of combining the two interventions; for example, GBG may increase attention to task and reduce disruptive behavior in the classroom, which in turn may facilitate the acquisition of the emotion regulation, social problem-solving, and conflict resolution skills taught in PATHS.
Methods:
We present the results of a 27-school RCT of the combined PAX GBG and PATHS programs, which is referred to as PATHS to PAX. Each of the 27 elementary schools (grades K-5) was assigned to one of 3 conditions: 1) Comparison/Control; 2) PAX GBG Alone; or 3) PATHS to PAX (PAX GBG+PATHS). We hypothesized that, relative to the control condition, both interventions would result in significantly lower levels of aggressive-disruptive and off-task behaviors at post-test. Moreover, we expected the effects of PATHS to PAX would be significantly greater than the PAX GBG condition alone. Classroom observational data and teacher ratings of the student behavior were collected at pre-test and at an end-of-year posttest. The sample included 5611 students enrolled in the urban project schools (50.81% male, 87.93% African American, and 85.96% low-SES).
Results:
Relative to the control condition at post-test, the PATHS to PAX condition manifested a significantly higher score on the On-Task Behavior scale and a significantly lower score on the Total Problem Behavior scale. Despite the larger number of significant PATH to PAX intervention effects, the contrasts between the PATHS to PAX and PAX GBG alone conditions only yielded significant differences for 2 out of the 7 comparisons.
Conclusions:
We found limited evidence to support our hypothesis of the superiority of the PATHS to PAX condition. However, we did find greater diversity of effects on the targeted risk behaviors but not necessarily their magnitude. With 2 exceptions, all the significant intervention effects consisted of intervention x pre-test level interactions, which is typical in universal preventive intervention trials (Ialongo et al., 1999). Additional results and implications of these findings will be discussed.