For those aiming to improve the social competence of others, the choice is usually between shaping specific micro-skills or more general problem-solving. While these are not mutually exclusive in practice, research has tended to investigate them separately. One of the major issues facing those targeting micro-skills is their validity in the setting of concern. For those targeting problemsolving, the issue is predominantly one of triggering problem-solving behaviour in the absence of trainer prompts. Other major issues which still have to be satisfactorily resolved include the identification and assessment of children requiring improved social skills and social competence. In addition, there is a need to remind ourselves that while improved social skills should lead to improved judgments about an individual's social competence, the latter is a somewhat abstract notion which tells us little about an individual's goals and objectives. While improved social competence should be one of our goals, it must be anchored to well defined behavioural objectives. One can be the life and soul of a party and receive high social competence ratings, yet be lonely and isolated once the party ends.

The present paper reviews recent developments in the assessment of children's social competence and social skills. Current conceptual models of social competence and social skills are discussed and provide a basis against which assessment measures can be evaluated. Although many measures are now available, there is a lack of integration of assessment devices within a conceptually sound framework of an assessment system. Such an assessment system should distinguish between social competence and social skills, should take into account developmental, sex and cultural factors, and should assess at micro-versus macro-skill and overt versus cognitive levels of behaviour. The system should then integrate information within a cognitive-behavioural analysis, using data from a wide range of informants, from a range of settings. Greater attention to matters of reliability and validity of assessment measures is also necessary in the development of such a system.

The aim of this study was to investigate the feelings of loneliness among students with learning disabilities (LD) and to identify subgroups according to their loneliness characteristics. The sample consisted of 76 students from seven self-contained classes for LD students within regular schools in all parts of Israel. The instruments included self-reported loneliness and social skills scales, teacher ratings of behaviour disorders, and peer ratings of social acceptance. The following factors predicted the students' feelings of loneliness: peer acceptance, social skills, and computer activities, explaining 32% of the variance. Using cluster analysis for the loneliness and behaviour disorder factors, four subgroups of students were identified. They differed along their levels of loneliness, presence of disruptive behaviour (internal validation) and social skills (external validation). The results emphasised the need to consider loneliness levels and manifestations of disruptive behaviour in attempts to conceptualise these students' difficulties and to plan individualised intervention programs. Further research, focusing on the impact of the intervention on the subjective experience of loneliness, may advance our understanding of the students' needs.

We evaluated the empirical validation model of target selection for two boys with interpersonal problems. Behaviours previously identified as distinguishing popular from rejected boys on a role-play measure of social behaviour were targeted for change in these boys who had histories of peer rejection. We trained three of these empirically-derived behaviours — appropriate body orientation, adequate speech intonation, and providing reasons for denying unreasonable requests — in a multiple-baseline-across-behaviours design. On a role-play measure of social behaviour, improvements in the targeted behaviours occurred upon, but not before, introduction of the intervention, confirming the controlling effects of the intervention. Behavioural improvements generalised to untrained role-play scenes and to a novel role-play partner, and were maintained for two of the three behaviours at 4- and 6-month follow-up assessments. The boys also showed improvements in parent-reported child behaviour from pre- to post-training. The study demonstrates use of an empirical approach to the selection of treatment targets for children.

A rationale is provided for the development of a small group instructional approach to improving the competence of adolescents who reportedly experience difficulty making and maintaining friendships. Social problem-solving studies are briefly examined to provide support for such an approach, which is based on incorporating gradated in vivo assignments into the training process and using progress in these assignments as feedback for further assignment setting. The program, named STAMPS (Social Training for Adolescents: Making Positive Steps) is described in some detail and illustrated with data from a single case. Discussion deals with aspects of the program requiring further investigation, including the selection of peers as potential friends, and the need to modify the program to deal with broader adolescent issues including conflict with authority, teachers, and parents.