Assessing and enhancing generalization and social validity of social-skills interventions with children and adolescents.

The authors read every social-skills study they could find. They saw the same hole: kids used the skill in the clinic, then lost it at recess.

The paper tells us to stop hoping generalization will just happen. Instead, build it into the teaching plan and check it with cold, hard data.

Social validity was mostly a smiley-face questionnaire. The team wanted numbers—how often the kid actually gets invited to birthday parties.

Without clear rules, each study measured success differently, so no one could tell what really worked.

Winett et al. (1991) said the same thing two years earlier: social validity must show real-world impact, not happy parent comments. The 1993 paper adds the "how-to" checklist.

Hong et al. (2018) ran the numbers the target paper asked for. Their meta-analysis shows caregiver coaching helps, but generalization still lags behind maintenance—exactly the gap the target warned about.

Hutchins et al. (2020) found the same split in schools: maintenance effects were almost twice as strong as generalization effects. Three decades later, the problem the target spotted is still alive.

Mann et al. (2020) gives a happy example: they taught college students conversation skills, then tracked generalization to untrained professors and collected social-validity data the target way—proof the blueprint works when you follow it.

Next time you write a social-skills goal, add a generalization plan and a social-validity probe. Measure invites, not smiles. If the skill doesn’t survive the lunchroom, it isn’t mastered.