Introduction: The aim of the present research was to conduct a meta-analysis on the effectiveness of play therapy studies on behavioral disorders in children.Method: The method of this study was meta-analysis. Twenty two studies from throughout Iran (1996 to 2012) including 637 participants were used in this meta-analysis. Data was analyzed using standardized mean difference effect size method for fixed effect model or Cohen’s d (Cohen, 1988, Hooman, 2008). All statistical operations were accomplished by Comprehensive Meta-Analysis, Version 2.0 (Borenstein, Hedges, Higgins & Rothestein, 2005, translated by Delavar & Ganji, 2012).Results: Results showed that the mean of mixed-model of effect size was 0/434 for the fixed effect model and 0/477 for the random effect model that both of them were significant in 0/001 level. In view of the fact that the effect sizes were heterogeneous, searches were done for moderator variables. Results showed that there wasn’t any significant difference between the effect sizes of studies, regarding the research design and gender of participants. The differences between the effect sizes of studies about the effectiveness of play therapy studies on behavioral disorders in children, regarding the age of participants and number of play therapy sessions were significant.Conclusion: It can be concluded that carrying out play therapies in many sessions and in lower ages could decrease behavioral disorder symptoms in children.