Translated from the Portuguese original. Read the original (Português)
Instead of arranging one thing with each child, the teacher sets a single rule for the whole class: “if this happens, this consequence follows.” That arrangement is a group contingency, and it can be built in three quite different ways — each student answering only for what they did (independent), everyone depending on the collective outcome (interdependent), or the whole group depending on one classmate’s performance (dependent). It is a cheap, practical tool with a solid research base, especially in the form of the Good Behavior Game. It is also the arrangement that demands the most ethical care: poorly planned, it turns a classmate into a scapegoat.
A group contingency (or group-oriented contingency) is an arrangement in which a single contingency — the same response requirement and the same consequence — is in effect for a set of people, rather than being programmed individually for each one. What defines the type of arrangement is not the fact that the procedure is applied “to a group,” but whose behavior produces the consequence and who receives it. Litow and Pumroy (1975) organized these systems into three categories — independent, interdependent, and dependent — a classification that remains the field’s reference.
Same rule, three different arrangements
Picture the same deal announced in three classrooms: “whoever finishes the reading task with 80% correct gets fifteen minutes of game time.” It sounds like a single rule, but what happens next depends on a detail that is easy to miss — who has to meet the criterion for someone to receive the reward. That detail is exactly what separates the three types.
Tap the cards to see how each arrangement works in practice — and what it tends to produce:
It is worth keeping a distinction that confuses a lot of people: applying a procedure to a group does not make it a group contingency in the technical sense. Thirty children each receiving their own individual token economy, side by side in the same room, remain under thirty separate contingencies. What characterizes the arrangement is the link between some people’s behavior and everyone’s consequence.
The Good Behavior Game
The most studied example of a group contingency was born in a fourth-grade classroom in 1969. Barrish, Saunders, and Wolf split the class into two teams “to play a game.” Two simple rules: don’t leave your seat and don’t talk out of turn. Each such episode earned a mark on the board for the student’s team. The team that ended below the agreed limit won privileges already available at the school — extra recess, being first to leave for lunch, time for special projects, stars and badges. The most important detail: both teams could win at the same time. It wasn’t a competition where someone had to lose. Leaving one’s seat and talking out of turn dropped sharply.
Hence the title of the original article, which sums up the mechanism well: individual contingencies for group consequences. Each student answers for what they do, but the result reaches the team — an interdependent arrangement. The game spread because it is cheap, requires little training, and uses reinforcers the school already has.
- Few, observable rules. “Talking out of turn” can be counted; “being respectful” cannot. It is worth revisiting the entry on operational definition.
- Teams, not individuals. The score belongs to the team, which dilutes the exposure of whoever makes the most mistakes.
- An achievable criterion. The limit on marks needs to be possible for the class you have today, not for the ideal class.
- More than one winner. When every team can win, the game stops rewarding anyone who roots for someone else’s mistake.
A positive variation became known as the Caught Being Good Game. Instead of marking violations, the teacher marks points for desired behaviors that they observe. The logic of the contingency is the same; what changes is what the adult ends up looking for in the room — and, with it, the climate of the procedure. It is an adjustment consistent with what is known about positive reinforcement and with the preference, in ethical practice, for procedures that build repertoire rather than just suppressing it (Wright & McCurdy, 2012).
What the research shows — and where it is more modestIn single-case studies, the game’s effects are consistent: a meta-analysis of 21 studies with 1,580 students, from preschool through high school, found a strong effect on behavior (TauU = 0.82; Bowman-Perrott et al., 2016), and another review of 22 studies estimated a medium effect (d ≈ 0.50; Flower et al., 2014). In a replication across five preschool classrooms, disruptive behavior dropped from averages of 13, 5, 4, 7, and 8 responses per minute to 2, 1, 1, 2, and 1 (Donaldson et al., 2011). The meta-analysis restricted to randomized trials — 7 studies, about 4,700 children — found much smaller effects (g between 0.09 and 0.32): there was a difference favoring the game for conduct problems (in both peer and teacher ratings) and for withdrawn behavior in peer ratings, but not for inattention or for teacher-rated withdrawn behavior (Smith et al., 2021). The honest reading is this: it works, it is very useful for classroom management, and it is not the transformative intervention the headlines suggest.
There is also a more ambitious chapter in this literature. Starting in the 1985–86 school year, in Baltimore, first- and second-grade classrooms were randomized to play the game or not, and the students were followed for more than a decade. At follow-up, at ages 19–21, the group that had played showed lower rates of drug and alcohol use disorders, regular smoking, antisocial personality disorder, and suicidal ideation (Kellam et al., 2008). Two cautions before turning this into a promise: the effects were concentrated among boys — especially those who were more aggressive in first grade — and barely appeared among girls; and this is a study in one city, with specific cohorts. It is a remarkable result, not an individual guarantee.
The delicate side: when the group turns into pressure
Tying the consequence for many to the behavior of some produces a predictable side effect: classmates start taking an interest in what each one does. That can be great — and it can go wrong. Explore the three points that come up most in the literature and in practice:
Blame, hostility, and sabotage
Four problems are documented in interdependent arrangements. The first is hostility toward the classmate who “made the group lose” — and when that classmate is always the same one, the procedure becomes an amplifier of exclusion. The second is sabotage: a student may discover that ruining the group’s chance is the most efficient way to get attention. The third is quieter: students who would work better stop at the minimum criterion, because that is enough for the reward to come. The fourth is giving up — once the limit has already been exceeded, there is nothing left to preserve for the rest of the day.
There is also a risk in how the data is read. A collective criterion like “the class average” can hide the student who did not improve — or who got worse — behind a nice-looking aggregate result. When there is an individual concern, it needs its own data collection.
Adjustments that change the outcome
- Randomize the components. Drawing which behavior, which criterion, and which reward count — revealing them only afterward — is the most studied solution. It reduces sabotage and working at the minimum, because no one knows what is being required or when it is already “not worth it anymore” (Popkin & Skinner, 2003).
- Keep the target anonymous in dependent arrangements: if the group does not know whose performance the reward depends on, the pressure has no address.
- Set the criterion against the baseline. An impossible limit guarantees failure; it is worth checking the baseline before choosing the number.
- An explicit rule against pressuring classmates. Hostility, threats, or blame are not part of the game — and that needs to be stated, not assumed.
- Never take back what has already been earned. A reward won is won; the next period starts from zero.
- Offer an individual alternative for whoever still cannot reach the collective criterion, running alongside the class system.
The arrangement is not just for school
Group contingencies show up in many other settings. In workplaces, bonuses for team goals and collective scoreboards are interdependent contingencies — and the same side effects tend to reappear there too, including peer pressure and effort calibrated to the bare minimum. In health programs, collective goals have been used to increase children’s physical activity in the classroom (Kuhl et al., 2015). At home, “if everyone is ready by eight, we’ll stop for ice cream” is the same structure.
One practical finding deserves note: in a study with teachers applying the game in a restrictive setting, there was a substantial reduction in problem behavior even with low implementation fidelity (Joslyn & Vollmer, 2020). That does not license applying it carelessly — but it suggests some tolerance for imperfect implementation, which helps explain the procedure’s popularity.
A group contingency is not collective punishmentThe difference is not one of degree, it is one of kind. Collective punishment penalizes everyone for one person’s act, without a prior agreement, without a clear criterion, and without a path to making things right. A well-built group contingency is agreed on beforehand, has an achievable and explicit criterion, delivers positive consequences, and plans for what happens when the criterion is not met. If, in practice, the outcome is one student being held responsible for the group’s frustration, the procedure is wrong — and the fix is to change the arrangement, not to demand more from the child. This is a matter of professional ethics, not a matter of style preference.
Test your understanding
Quick quiz
Three questions to lock in the concept. You see the explanation right away.
Further reading
- Litow, L.; Pumroy, D. K. “A brief review of classroom group-oriented contingencies”. Journal of Applied Behavior Analysis, 8, 341–347, 1975.
- Barrish, H. H.; Saunders, M.; Wolf, M. M. “Good behavior game: effects of individual contingencies for group consequences on disruptive behavior in a classroom”. Journal of Applied Behavior Analysis, 2, 119–124, 1969.
- Cooper, J. O.; Heron, T. E.; Heward, W. L. Applied Behavior Analysis. 3rd ed. Pearson, 2020 — chap. 28, “Token Economy, Group Contingencies, and Contingency Contracting”.
- Joslyn, P. R.; Donaldson, J. M.; Austin, J. L.; Vollmer, T. R. “The Good Behavior Game: a brief review”. Journal of Applied Behavior Analysis, 52, 811–815, 2019.
- Donaldson, J. M.; Vollmer, T. R.; Krous, T.; Downs, S.; Berard, K. P. “An evaluation of the Good Behavior Game in kindergarten classrooms”. Journal of Applied Behavior Analysis, 44, 605–609, 2011.
- Bowman-Perrott, L.; Burke, M. D.; Zaini, S.; Zhang, N.; Vannest, K. “Promoting positive behavior using the Good Behavior Game: a meta-analysis of single-case research”. Journal of Positive Behavior Interventions, 18, 180–190, 2016.
- Flower, A.; McKenna, J. W.; Bunuan, R. L.; Muething, C. S.; Vega, R. “Effects of the Good Behavior Game on challenging behaviors in school settings”. Review of Educational Research, 84, 546–571, 2014.
- Smith, S.; Barajas, K.; Ellis, B.; Moore, C.; McCauley, S.; Reichow, B. “A meta-analytic review of randomized controlled trials of the Good Behavior Game”. Behavior Modification, 45, 641–666, 2021.
- Kellam, S. G. et al. “Effects of a universal classroom behavior management program in first and second grades on young adult behavioral, psychiatric, and social outcomes”. Drug and Alcohol Dependence, 95 (suppl. 1), S5–S28, 2008.
- Wright, R. A.; McCurdy, B. L. “Class-wide positive behavior support and group contingencies: examining a positive variation of the Good Behavior Game”. Journal of Positive Behavior Interventions, 14, 173–180, 2012.
- Popkin, J.; Skinner, C. H. “Enhancing academic performance in a classroom serving students with serious emotional disturbance: interdependent group contingencies with randomly selected components”. School Psychology Review, 32, 271–284, 2003.
- Joslyn, P. R.; Vollmer, T. R. “Efficacy of teacher-implemented Good Behavior Game despite low treatment integrity”. Journal of Applied Behavior Analysis, 53, 465–474, 2020.
- Kuhl, S.; Rudrud, E. H.; Witts, B. N.; Schulze, K. A. “Classroom-based interdependent group contingencies increase children’s physical activity”. Journal of Applied Behavior Analysis, 48, 2015.
This entry is educational and does not replace assessment, diagnosis or treatment by a qualified health professional. Every person is unique — for a specific case, seek individual guidance.