Abstract
In a hospital training unit for severely mentally handicapped adults, alternative methods of staff organization, Room Management (RM) and Small Groups (SG) were experimentally evaluated. Both procedures were found to be equivalent in ensuring that trainees had materials available to them, in promoting on-task behaviour and staff trainee interaction and minimizing staff–staff interaction. Group engagement was significantly higher during the SG condition. Room Management was nevertheless significantly more successful in facilitating individual training than the SG arrangement, trainees receiving four times more individual training during RM than SG. The implications of these results for the appropriate use of training staff are discussed. The use of group engagement as a dependent variable is critically evaluated.