Using the Good Behavior Game

To start teaching behavior in my classroom, I decided to start the process of using the Good Behavior Game (GBG). I took scaffolded steps before a full-blown implementation. I explicitly taught behavior, collected data on frequent negative behaviors in my classroom, and found out what would rewards would motivate my students. I used the following resources to help me set up my own classroom implementation of the GBG as well as a data collection method:

1. Good behavior game: effects of individual contingencies for group consequences on disruptive behavior in a classroom (the original from 1967!)
2. Fallon, L. M., Marcotte, A. M., & Ferron, J. M. (2020). Measuring Academic Output During the Good Behavior Game: A Single Case Design Study. Journal of Positive Behavior Interventions, 22(4), 246–258. https://doi.org/10.1177/1098300719872778
3. Intervention Central: Good Behavior Game
4. Behavior Code by Jessica Minahan

In our school district, teacher evaluations consist of an option to conduct action-research. So, I implemented the GBG as part of my action-research project.

To start, I focused mainly on independent work times, which is where the most problem behavior occurred in the classroom. After collecting data on an ABC chart, I noted the following behaviors:

1. Vocal noises: As a class, we defined this together as any noise (words or sounds) that came from our throat and mouth. *We had to be clear on this. At the beginning, I initially only focused on talking as a form of noise. I found that my students then became creative with noise-making. Let’s just say that a group of 8 and 9 year olds taught me a bit more about operational definitions!
2. Leaving seats: Since students occasionally had to pick pencils off of the floor or grab materials, this was defined as a bum leaving a seat for longer than 10 seconds.

To turn these into positive statements, we worked together as a class to create the following expectations:

1. We will work quietly (no talking or vocal noises).
2. We will stay seated (not leaving seat for longer than 10 seconds).
3. We can talk or leave our seat with permission. Permission is given by a student first raising their hand above shoulder height.

The study in my own classroom was not without its limitations. First, I was the only one collecting data. Our school, like many, are experiencing a staffing shortage. Finding an extra person to help collect data or compare data with me is near impossible. There were even a couple of times when I would start the GBG, just to have to take care of a situation or have something come up that would keep me from continuing. I did get enough data to show its effectiveness.

Data was collected in three phases. The first phase was on a paper that the students could not see and was used as a baseline. This was done after the students were taught the behavior expectations, but without GBG implementation. Then, the GBG was implemented and data was collected on the board for the students to see. The last phase was a reversal and data was collected in the same manner as the baseline.

The GBG was played in 10-15 minute intervals. Students were told the behavior expectations each time. The behavior expectations were posted on the board as a reminder as well as what could be earned for having the expected behavior. The rewards were chosen based on feedback from the Forced Choice Inventory as well as student input.

Students were then sectioned into groups based on classroom seating. Tallies were given discretely (no verbal redirection or public embarrassment, just a quiet tally) when a student did not follow that expectation. For example, if a student was talking to a partner, they received a tally for that behavior since it did not meet the quiet expectation.

The tallies were then totaled for the number of minutes the GBG was played. For graphing purposes, the tallies were converted to number of behavior occurrences per minute.

The first phase shows the number of behavior occurrences (talking, vocal noises, leaving seat for more than 10 seconds… all without permission gained by raising hand) per minute without the implementation of the GBG. During the baseline phase, the defined negative behavior occurred an average of 1.10 times per minute. With the implementation of the GBG, the defined negative behavior reduced to an average of 0.27 times per minute. In the reversal phase, the average number of occurrences of negative behavior increased to an average of 0.63 times per minute. Now, I know that best practice is to have at least three data points per phase, but time and resources were not on my side for the reversal at the end. If I was creating a published study, I would have ensured that there was another data collection point for the reversal. However, I was very pleased with the reduction in negative behavior and the increase in expected behavior.

Here’s the deal – The Good Behavior Game works! It has been studied since 1967 and I have not come across a study yet to prove other wise. I know that in my own classroom, students beg me to play the GBG. When I do, so much more work is accomplished and students are focused on the task at hand.

My next classroom study is to implement a Random Interval Timer. I am still creating a plan and working on how to collect data using this method. Let me know if you have tried the GBG or Random Interval Timers in your setting!