Generalization in social behavior research with children and youth who have emotional or behavioral disorders.

The authors read every social-skills study they could find on students with emotional or behavioral disorders.

They looked for two things: did the researchers test if the new skills showed up in new places, and did they keep track of them over time.

Most papers came up short on both counts.

Hardly any studies planned real-world probes. Most stopped testing once the child mastered the skill at a desk.

Maintenance data were even thinner; long-term follow-up was rare.

Gianoumis et al. (2012) show the fix is already on the shelf. In staff-training reviews, common stimuli, multiple examples, and self-cue cards pop up again and again.

Neely et al. (2018) echo the worry in functional-communication reviews: only six of thirty-seven studies gave solid generalization proof.

McGee et al. (1983) proves the point with kids, not just staff. Six behaviorally-handicapped students used self-evaluation cards to keep good behavior alive when they moved from resource room to gen-ed.

The 1992 paper scolds the field for ignoring these tactics; the later reviews show the scolding still applies.

If you write a behavior plan without a generalization line, you are planning for failure. Add a probe in the lunchroom, a data point next month, and a self-check card the student can carry. These small moves turn clinic success into real-life skill.