The thesis evaluated the effectiveness of a school-based social competence promotion programme for pupils aged 9 to 11. At present, most of the psychosocial health promotion programmes aim at a specific risk factor (e.g. low social competence) but do not consider possible interactions with other risk factors like gender, ethnicity or socio-economic status, which may influence the effectiveness of these programmes. The main research question was, whether a non subgroup specific social competence promotion could be equally effective for all participating pupils. Two causal models have been developed to test this question empirically. The first model tested the effectiveness of the programme for the whole sample, the second model examined the effects for pupils with a low, a middle or a high socio-economic status.
The evaluation was designed as a quasi-experimental study with a pretest-posttest-posttest design, using a questionnaire for the data collection. The sample contained 716 pupils of which 49 p.c. belonged to the intervention group. Nearly 80 p.c. of the pupils were German, with Turkish pupils as the largest minority group. The tests of the models were conducted with the method of Structural Equations of Latent Variables in a Panel Design.
The empirical results showed that the second model gives a better representation of the effects of the programme. The programme was able to improve the social competences for children with a high socio-economic status directly after the end of the intervention. For children with a low socio-economic status the programme seemed to enhance the awareness of lack of competence but it was not able to improve the social competences in this group. There was no effect found for children with a middle socio-economic status.