Validation of the School-age Assessment of Attachment in a short-term longitudinal study.

Abstract

This study tested the validity of the School-aged Assessment of Attachment (SAA) in terms of matches from the well-validated Preschool Assessment of Attachment (PAA) to the SAA 6 months later. It also addressed validity in terms of mental health services and measures of stress, depression and anxiety. Children's SAA classifications were predicted to match their 6 months' previous PAA classifications and indicators of maternal, child and family stress. The study used a two-group comparative design, involving normative and clinical children and their mothers. The participants were 50 children between 5.5 and 5.9 years of age. Each child participated with his or her mother in a PAA, and then 6 months later each child responded to the SAA story cards as well as self-report assessments of stress, anxiety and depression. Concordance of A, B, C and A/C attachment classifications was found between the PAA and SAA in 34 of 48 children. There was a strong relation between referral status (clinical or normative) and both PAA and SAA attachment classifications. In every non-matching case, a normative child had an attachment classification indicative of risk, indicating that the direction of errors was false positives as opposed to false negatives. This evidence supports the validity and clinical utility of the SAA.