All studies were rated for quality using commonly accepted standards for group comparison research methods. The purpose of these ratings was to create indicators of the overall confidence that can reasonably be placed in the studies. The quality ratings ranged from a low rating of 4.7 to a high rating of 9.7. The average quality rating was 7.04 with a median rating of 6.75 out of a possible score of 10. The most common design deficiencies were lack of methods for assuring fidelity of implementation (12 of 16), lack of treatment manuals (9 of 16 without), and lack of follow-up data (9 of 16). Several studies failed to report important descriptive data on the parents and children.