Positive reinforcement is not as easy as it looks, according to a recent study of Maryland elementary schools that implemented Positive Behavioral Interventions and Supports (PBIS).
PBIS seeks to improve student behavior with a 3-tiered system that first aims to prevent problem behaviors by clearly defining and communicating behavioral expectations for all students and rewarding behavior that meets those expectations.
Surprisingly and not so surprisingly, while schools may be more concerned about students who need secondary and tertiary interventions, the team of researchers from Johns Hopkins University found that schools need more help in teaching and defining behavioral expectations to all students than in responding to behavioral violations.
This may be because, traditionally, schools have taken a more reactive and punitive approach to student behavior and have more resources for dealing with violations, but are less equipped for promoting desired behavior, says the article recently published in Education and Treatment of Children.
“We recommend that a significant portion of the initial training in PBIS focus on strategies for teaching the behavioral expectations, including the development of written lesson plans,” the researchers write. Coaches who work with schools report that schools find it challenging to develop lessons for teaching behavior expectations, the researchers say. One of the primary objectives of PBIS training, should be to encourage a shift from punishment-based strategies to teaching and reinforcing prosocial behaviors, the authors write.
Students who do not respond to the primary prevention effort receive secondary interventions that provide group support to encourage desirable behaviors (e.g., programs that teach organization or social skills). For students who do not respond to the secondary group intervention, an individualized tertiary intervention then focuses on promoting the desirable behavior while extinguishing the student’s undesirable behavior.
Schools may already be practicing some aspects of PBIS, for example, in the way they respond to students with problem behaviors, the researchers write. It is important to take an inventory of what schools are already doing before implementing the system so that there is no duplication of effort, they say.
Importance of good baseline data
“This baseline data could then be used by administrators, PBIS behavior support coaches, and PBIS trainers to identify the strengths and weaknesses specific to the school,” the researchers write. “Trainings could then be tailored to target the needs of the school.”
Of the 37 elementary schools that participated in the study, 21 received training in implementing PBIS and 16 did not receive training. The schools that did receive training formed internal PBIS teams comprising 5-6 members. These teams attended a 2-day summer training led by the developers of PBIS, George Sugai and Rob Horner. Local coaches, a regional PBIS coordinator and the Maryland State PBIS Leadership Team provided ongoing support. School team members also attended 2-day summer “booster” training events in subsequent years.
To evaluate fidelity to PBIS implementation, trained assessors evaluated all schools using the School-Wide Evaluation Tool (SET). SET consists of 29 items organized into the following seven subscales:
- expectations defined;
- behavioral expectations taught;
- system for rewarding behavioral expectations;
- system for responding to behavioral violations;
- monitoring and evaluation;
- management; and
- district-level support
The researchers say trained schools scored significantly higher on implementation fidelity to PBIS than non-trained schools. Specifically, 67% (14 out of 21) of the trained schools achieved the 80% implementation level on the total SET score by the end of the first year. By contrast only one of the 16 non-trained schools scored above 80% at the end of the first and year and 3 by the end of the second year. Most scored between 30% and 50% on SET
“Although the developers of PBIS have theorized that it takes three to five years to implement the model, these data suggest that most schools were able to implement the program with high fidelity in just one to two years,” the authors write. “However, it may take the full three to five years for these organizational changes to translate into changes in student behavior, such as reduced behavior problems and increased academic performance.”
Schools implementing PBIS should conduct frequent SET evaluations (e.g. monthly, quarterly, bi-annually) to better monitor their progress, the researchers say. Many initiatives get off to a great start but then falter on the build-up and maintenance. Frequent SET evaluations make it possible to give school personnel more feedback and to decrease the amount of time it takes for schools to reach thresholds for effective implementation. School personnel can receive copies of SET data which can provide guidance in changes needed in current practices or areas where more training is needed.
Schools can, to a certain extent, self-train on positive behavior support strategies. However, the study found that training and coaching will likely result in quicker and more substantial gains in school-wide PBIS implementation.
Non-trained schools were more likely to implement the components of the model that are consistent with traditional behavioral discipline approaches, but less likely to teach positive behavioral expectations and reward children for meeting those expectations. In schools that did use a reward system for behaviors, fewer of them linked the rewards with postings or teaching school-wide behavioral expectations, the researchers write. “Thus, it is recommended that school personnel teach behavioral expectations to students in a similar way as they teach academics,” the authors write.
“Implementation of School-Wide Positive Behavioral Interventions and Supports (PBIS) in Elementary Schools: Observations from a Randomized Trial,” by Catherine Bradshaw, Wendy Reinke, et. al. Education and Treatment of Children, Volume 31, Number 1, 2008, pp. 1-26.