Measurement of Young Children's Close Friendship: A Comparison of Two Independent Assessment Systems and their Associations with Attachment Security

Abstract

The purpose of this study was to compare two independent dyadic assessments of children's close friendship and to examine the attachment correlates of both measures. A total of 73 5-year-olds, who had participated in a longitudinal study of child and family development with their parents and their close friend were observed in a 30 minute laboratory freeplay setting. Each friendship pair was: (1) rated every 30 seconds on eight dimensions of close relationships (e.g. connectedness, negativity, synchrony) using the Dyadic Coding System (DCS: Youngblade & Belsky, 1992); (2) sorted on seven similar dimensions of relationships (e.g. positive social orientation, harmony, cohesiveness) using the Dyadic Relationships Q-sort (DRQ: Park & Waters, 1989). Antecedent attachment data were collected at 12 (with mother) and 13 (with father) months in the Strange Situation; at this time each parent also completed the Attachment Q-sort (Waters & Deane, 1985). Each parent completed the Attachment Q-sort again at 36-37 months. The results revealed that both friendship measures captured similar variation in friendship quality. Analyses of the links between child-parent attachment and friendship suggested congruence between Q-sort attachment security and friendship quality measured with the DRQ, but only for the child-father relationship. Analyses using Strange Situation assessments of infant-father attachment revealed counterintuitive associations with friendship quality, as measured by both the DRQ and DCS. There were no statistically significant relations between child-mother attachment security and friendship quality. In general, the findings point to a number of complexities regarding the measurement and interpretation of links between social relationships.