A meta-analysis of 100 studies of outcomes of play therapy interventions was conducted to determine overall effect of intervention. Different from the Bratton, Ray, Rhine, and Jones (2005), but more consistent with the Lin and Bratton (2015), findings, there was a significant moderate effect for play therapy interventions across all outcomes (d = .44). When 4 studies with aberrantly large effects were removed, this effect was d = .36. These moderate effects are consistent with other meta-analyses that find lower effect sizes for nonbehavioral interventions. Overall study quality was poor with no studies meeting the criteria of randomized control trials. Research on play therapy interventions also does not use diagnostic criteria as part of study inclusion, thus preventing play therapy interventions from being considered as empirically supported treatments. Meta-analytic findings differed by type of measure used, with measures of family functioning/relationships finding larger results than other types of measures. There was not strong consistency in measures used across the studies, with only a few measures being used across more than 1 or 2 studies. Of the more frequently used measures, the Measurement of Empathy in Adult Child Interaction (MEACI) resulted in much larger effects than other measures used, and should be further evaluated in terms of appropriate interpretation and use. Effect sizes also differed based on the reporting source, with teachers tending to rate lower impacts of treatment than other reporters.
