The Good Behaviour Game is one of the most replicated classroom interventions in behaviour analysis. The original 1969 study — Barrish, Saunders and Wolf, listed in the evidence base below — has been followed by five decades of replication across ages, settings, and variations. This synthesis pulls that evidence together; the studies it draws on are listed at the end.

Variations

Researchers have tested many versions of the game. The headline is that most of them work — the choice is usually about fit and teacher effort rather than effectiveness.

  • Positive reinforcement vs. response cost. The “Caught Being Good Game” awards points for following the rules instead of marking rule-breaking. Both versions are about equally effective — though some students actually prefer the response-cost version, and a few may blame teammates for a loss.
  • Team structures. The traditional game uses two or more teams, but a “no-team” version where the whole class plays as one is also effective, lowers teacher effort, and suits high-school classes well.
  • Student-led games. Even early-elementary students can be trained to review the rules, score tallies, and deliver rewards. Student-led games are as effective as teacher-led ones, cut teacher workload, and are strongly preferred by students.
  • Feedback delivery. For older students, combining vocal feedback (naming the rule broken) with visual feedback (tally marks) works best. In preschool, vocal feedback appears to be necessary — visual tallies alone may not be enough to reduce disruption.
  • The reward is not optional. Stating rules and keeping score without a contingent reward does not produce lasting change. An external reward — free time, a privilege, a small item — is what drives it.
  • Technology. Digital versions using tools like ClassDojo are as effective as a whiteboard tally, and can appeal to students with autism.

How strong is the evidence

The evidence base is unusually strong for a classroom intervention: rigorous single-case designs (reversal and multiple-baseline) and meta-analyses of that single-case research, spanning five decades. The effect on disruptive behaviour is consistent and well established.

A few honest caveats:

  • It is a behaviour intervention, not an academic one. Improvements in academic output, such as writing fluency, are modest; the game needs to sit alongside strong instruction, not replace it.
  • Generalisation is weak. The effects largely hold only while the game is being played — it works more like a prosthetic environment during active play than a permanent change across the whole day.
  • Diminishing returns in the most severe settings. For students with severe psychiatric disorders in specialised settings, effects are only modest and may need pairing with targeted social-skills teaching.
  • Watch for negative peer pressure with the response-cost version, where a student can be blamed for a team’s loss.

Frequently Asked Questions

Does the Good Behaviour Game actually work?

Yes — it is one of the best-evidenced classroom interventions in behaviour analysis. Decades of single-case research, and meta-analyses of that research, show consistent moderate-to-large reductions in disruptive and off-task behaviour across ages from preschool to high school. It is also unusually forgiving: it keeps reducing disruption even when teachers score imperfectly.

Does the Good Behaviour Game improve academic achievement?

Its main, best-established effect is on behaviour — disruption and off-task behaviour fall and academic engagement rises. Gains in actual academic output, such as writing fluency, are more modest. Treat it as a way to create the conditions for learning alongside strong instruction, not as an academic intervention in its own right.

Do the effects of the Good Behaviour Game last?

The effects are immediate but tend to stay confined to the periods when the game is actually played — behaviour does not automatically improve in the lessons before or after. It works more like a supportive structure during active play than a permanent change across the whole day, so it is best run during your most challenging periods rather than expected to generalise on its own.

What is the Caught Being Good Game?

It is a positive-reinforcement version that awards points for following the rules instead of marking rule-breaking. The evidence suggests it is about as effective as the traditional response-cost version, and some students prefer it — though, like any version, it needs balanced teams and a generous criterion to avoid negative peer pressure when a team loses.

This page is the evidence base. For a step-by-step guide to actually running the game — building balanced teams, setting a criterion students can hit, and designing out the failure modes above — see The Good Behaviour Game: a whole-class strategy that actually has evidence.