The purpose of this study was to investigate the outcomes of social communication interventions for school-age students with Asperger syndrome and to investigate appropriate method for summarizing intervention outcomes for single-subject research. Five effect sizes(PND, PEM, ECL, PADN, NAP) were calculated for articles and correlations were calculated to determine the degree of relationship five effect sizes metrics. 23 articles were selected by the criteria for this synthesis and were analyzed across subjects', settings, and intervention characteristics and effect sizes were calculated by PND of the non-regression model. 14 articles were single subject research studies and 9 articles were group experimental studies. A total of 303 subjects were included, of whom 271 were boys and 32 were girls. The analyses of 14 single-subject research studies resulted in 107 PND scores. Results varied widely across subjects', settings, and intervention variables. Variables that lead to the most effective intervention by PND were upper-grade elementary subjects, AS subjects, special classroom settings, 15-20 sessions of interventions, interventions by teachers, multiple baseline design across settings, mind reading intervention, social behavioral learning strategy. There were significant very high correlation between the PAND and the NAP effect sizes, and was high correlation between metrics(PND, PEM, PAND, NAP) except ECL. Based on the result, the implication for education for children with Asperger syndrome was discussed, and future study issues were recommended.